Warning: Permanently added '2620:52:3:1:dead:beef:cafe:c152' (ED25519) to the list of known hosts. You can reproduce this build on your computer by running: sudo dnf install copr-rpmbuild /usr/bin/copr-rpmbuild --verbose --drop-resultdir --task-url https://copr.fedorainfracloud.org/backend/get-build-task/8472918-fedora-rawhide-x86_64 --chroot fedora-rawhide-x86_64 Version: 1.2 PID: 26287 Logging PID: 26288 Task: {'allow_user_ssh': False, 'appstream': False, 'background': False, 'build_id': 8472918, 'buildroot_pkgs': [], 'chroot': 'fedora-rawhide-x86_64', 'enable_net': False, 'fedora_review': False, 'git_hash': '447e2c2466aa37c15327fc380763d058772e3278', 'git_repo': 'https://copr-dist-git.fedorainfracloud.org/git/@rocm-packagers-sig/RH/llama-cpp', 'isolation': 'default', 'memory_reqs': 2048, 'package_name': 'llama-cpp', 'package_version': 'b4094-8', 'project_dirname': 'RH', 'project_name': 'RH', 'project_owner': '@rocm-packagers-sig', 'repo_priority': None, 'repos': [{'baseurl': 'https://download.copr.fedorainfracloud.org/results/@rocm-packagers-sig/RH/fedora-rawhide-x86_64/', 'id': 'copr_base', 'name': 'Copr repository', 'priority': None}], 'sandbox': '@rocm-packagers-sig/RH--trix', 'source_json': {}, 'source_type': None, 'ssh_public_keys': None, 'storage': 0, 'submitter': 'trix', 'tags': [], 'task_id': '8472918-fedora-rawhide-x86_64', 'timeout': 18000, 'uses_devel_repo': False, 'with_opts': [], 'without_opts': []} Running: git clone https://copr-dist-git.fedorainfracloud.org/git/@rocm-packagers-sig/RH/llama-cpp /var/lib/copr-rpmbuild/workspace/workdir-g_cwufk3/llama-cpp --depth 500 --no-single-branch --recursive cmd: ['git', 'clone', 'https://copr-dist-git.fedorainfracloud.org/git/@rocm-packagers-sig/RH/llama-cpp', '/var/lib/copr-rpmbuild/workspace/workdir-g_cwufk3/llama-cpp', '--depth', '500', '--no-single-branch', '--recursive'] cwd: . rc: 0 stdout: stderr: Cloning into '/var/lib/copr-rpmbuild/workspace/workdir-g_cwufk3/llama-cpp'... Running: git checkout 447e2c2466aa37c15327fc380763d058772e3278 -- cmd: ['git', 'checkout', '447e2c2466aa37c15327fc380763d058772e3278', '--'] cwd: /var/lib/copr-rpmbuild/workspace/workdir-g_cwufk3/llama-cpp rc: 0 stdout: stderr: Note: switching to '447e2c2466aa37c15327fc380763d058772e3278'. You are in 'detached HEAD' state. You can look around, make experimental changes and commit them, and you can discard any commits you make in this state without impacting any branches by switching back to a branch. If you want to create a new branch to retain commits you create, you may do so (now or later) by using -c with the switch command. Example: git switch -c Or undo this operation with: git switch - Turn off this advice by setting config variable advice.detachedHead to false HEAD is now at 447e2c2 automatic import of llama-cpp Running: dist-git-client sources 100 18.6M 100 18.6M 0 0 6834k 0 0:00:02 0:00:02 --:--:-- 6833k INFO: Reading stdout from command: md5sum llama.cpp-b4094.tar.gz /usr/bin/tail: /var/lib/copr-rpmbuild/main.log: file truncated Running (timeout=18000): unbuffer mock --spec /var/lib/copr-rpmbuild/workspace/workdir-g_cwufk3/llama-cpp/llama-cpp.spec --sources /var/lib/copr-rpmbuild/workspace/workdir-g_cwufk3/llama-cpp --resultdir /var/lib/copr-rpmbuild/results --uniqueext 1736126189.030111 -r /var/lib/copr-rpmbuild/results/configs/child.cfg INFO: mock.py version 6.0 starting (python version = 3.13.0, NVR = mock-6.0-1.fc41), args: /usr/libexec/mock/mock --spec /var/lib/copr-rpmbuild/workspace/workdir-g_cwufk3/llama-cpp/llama-cpp.spec --sources /var/lib/copr-rpmbuild/workspace/workdir-g_cwufk3/llama-cpp --resultdir /var/lib/copr-rpmbuild/results --uniqueext 1736126189.030111 -r /var/lib/copr-rpmbuild/results/configs/child.cfg Start(bootstrap): init plugins INFO: tmpfs initialized INFO: selinux enabled INFO: chroot_scan: initialized INFO: compress_logs: initialized Finish(bootstrap): init plugins Start: init plugins INFO: tmpfs initialized INFO: selinux enabled INFO: chroot_scan: initialized INFO: compress_logs: initialized Finish: init plugins INFO: Signal handler active Start: run INFO: Start(/var/lib/copr-rpmbuild/workspace/workdir-g_cwufk3/llama-cpp/llama-cpp.spec) Config(fedora-rawhide-x86_64) Start: clean chroot Finish: clean chroot Mock Version: 6.0 INFO: Mock Version: 6.0 Start(bootstrap): chroot init INFO: mounting tmpfs at /var/lib/mock/fedora-rawhide-x86_64-bootstrap-1736126189.030111/root. INFO: calling preinit hooks INFO: enabled root cache INFO: enabled package manager cache Start(bootstrap): cleaning package manager metadata Finish(bootstrap): cleaning package manager metadata INFO: Guessed host environment type: unknown INFO: Using container image: registry.fedoraproject.org/fedora:rawhide INFO: Pulling image: registry.fedoraproject.org/fedora:rawhide INFO: Tagging container image as mock-bootstrap-3f85572a-5305-4026-b810-0d9cb548fe2d INFO: Checking that ef25fceac2d8232ea43e7014e23599addde212ebff5f939734b9d30f7da20a73 image matches host's architecture INFO: Copy content of container ef25fceac2d8232ea43e7014e23599addde212ebff5f939734b9d30f7da20a73 to /var/lib/mock/fedora-rawhide-x86_64-bootstrap-1736126189.030111/root INFO: mounting ef25fceac2d8232ea43e7014e23599addde212ebff5f939734b9d30f7da20a73 with podman image mount INFO: image ef25fceac2d8232ea43e7014e23599addde212ebff5f939734b9d30f7da20a73 as /var/lib/containers/storage/overlay/ff1ae2f0e2b0f324dd1019713bf5404259e5a5fd53afc4de346b94abbab61867/merged INFO: umounting image ef25fceac2d8232ea43e7014e23599addde212ebff5f939734b9d30f7da20a73 (/var/lib/containers/storage/overlay/ff1ae2f0e2b0f324dd1019713bf5404259e5a5fd53afc4de346b94abbab61867/merged) with podman image umount INFO: Removing image mock-bootstrap-3f85572a-5305-4026-b810-0d9cb548fe2d INFO: Package manager dnf5 detected and used (fallback) INFO: Not updating bootstrap chroot, bootstrap_image_ready=True Start(bootstrap): creating root cache Finish(bootstrap): creating root cache Finish(bootstrap): chroot init Start: chroot init INFO: mounting tmpfs at /var/lib/mock/fedora-rawhide-x86_64-1736126189.030111/root. INFO: calling preinit hooks INFO: enabled root cache INFO: enabled package manager cache Start: cleaning package manager metadata Finish: cleaning package manager metadata INFO: enabled HW Info plugin INFO: Package manager dnf5 detected and used (direct choice) INFO: Buildroot is handled by package management downloaded with a bootstrap image: rpm-4.20.0-1.fc42.x86_64 rpm-sequoia-1.7.0-3.fc42.x86_64 dnf5-5.2.8.1-2.fc42.x86_64 dnf5-plugins-5.2.8.1-2.fc42.x86_64 Start: installing minimal buildroot with dnf5 Updating and loading repositories: fedora 100% | 747.3 KiB/s | 25.4 KiB | 00m00s Copr repository 100% | 31.2 KiB/s | 1.5 KiB | 00m00s Copr repository 100% | 1.4 MiB/s | 120.0 KiB | 00m00s Repositories loaded. Package Arch Version Repository Size Installing group/module packages: bash x86_64 5.2.37-1.fc42 fedora 8.2 MiB bzip2 x86_64 1.0.8-19.fc41 fedora 95.7 KiB coreutils x86_64 9.5-11.fc42 fedora 5.4 MiB cpio x86_64 2.15-2.fc41 fedora 1.1 MiB diffutils x86_64 3.10-8.fc41 fedora 1.6 MiB fedora-release-common noarch 42-0.11 fedora 19.8 KiB findutils x86_64 1:4.10.0-4.fc41 fedora 1.8 MiB gawk x86_64 5.3.0-4.fc41 fedora 1.7 MiB glibc-minimal-langpack x86_64 2.40.9000-26.fc42 fedora 0.0 B grep x86_64 3.11-9.fc41 fedora 1.0 MiB gzip x86_64 1.13-2.fc41 fedora 389.0 KiB info x86_64 7.1.1-2.fc42 fedora 361.8 KiB patch x86_64 2.7.6-25.fc41 fedora 266.7 KiB redhat-rpm-config noarch 296-1.fc42 fedora 186.6 KiB rpm-build x86_64 4.20.0-1.fc42 fedora 194.3 KiB sed x86_64 4.9-3.fc41 fedora 861.5 KiB shadow-utils x86_64 2:4.17.0-2.fc42 fedora 4.0 MiB tar x86_64 2:1.35-4.fc41 fedora 2.9 MiB unzip x86_64 6.0-65.fc42 fedora 398.2 KiB util-linux x86_64 2.40.2-8.fc42 fedora 3.7 MiB which x86_64 2.21-42.fc41 fedora 80.2 KiB xz x86_64 1:5.6.3-2.fc42 fedora 1.2 MiB Installing dependencies: add-determinism x86_64 0.5.0-1.fc42 fedora 2.4 MiB alternatives x86_64 1.31-1.fc42 fedora 64.8 KiB ansible-srpm-macros noarch 1-16.fc41 fedora 35.7 KiB audit-libs x86_64 4.0.2-1.fc41 fedora 331.3 KiB authselect x86_64 1.5.0-8.fc42 fedora 157.5 KiB authselect-libs x86_64 1.5.0-8.fc42 fedora 822.2 KiB basesystem noarch 11-21.fc41 fedora 0.0 B binutils x86_64 2.43.50-9.fc42 fedora 25.8 MiB build-reproducibility-srpm-macros noarch 0.5.0-1.fc42 fedora 735.0 B bzip2-libs x86_64 1.0.8-19.fc41 fedora 80.7 KiB ca-certificates noarch 2024.2.69_v8.0.401-4.fc42 fedora 2.6 MiB coreutils-common x86_64 9.5-11.fc42 fedora 11.2 MiB cracklib x86_64 2.9.11-6.fc41 fedora 238.9 KiB crypto-policies noarch 20241128-1.gitbb7b0b0.fc42 fedora 137.3 KiB curl x86_64 8.11.1-2.fc42 fedora 452.0 KiB cyrus-sasl-lib x86_64 2.1.28-27.fc41 fedora 2.3 MiB debugedit x86_64 5.1-2.fc42 fedora 200.3 KiB dwz x86_64 0.15-8.fc42 fedora 299.2 KiB ed x86_64 1.20.2-2.fc41 fedora 146.9 KiB efi-srpm-macros noarch 5-13.fc42 fedora 40.2 KiB elfutils x86_64 0.192-7.fc42 fedora 2.6 MiB elfutils-debuginfod-client x86_64 0.192-7.fc42 fedora 81.4 KiB elfutils-default-yama-scope noarch 0.192-7.fc42 fedora 1.8 KiB elfutils-libelf x86_64 0.192-7.fc42 fedora 1.2 MiB elfutils-libs x86_64 0.192-7.fc42 fedora 662.9 KiB fedora-gpg-keys noarch 42-0.3 fedora 126.4 KiB fedora-release noarch 42-0.11 fedora 0.0 B fedora-release-identity-basic noarch 42-0.11 fedora 719.0 B fedora-repos noarch 42-0.3 fedora 4.9 KiB fedora-repos-rawhide noarch 42-0.3 fedora 2.2 KiB file x86_64 5.45-8.fc42 fedora 103.7 KiB file-libs x86_64 5.45-8.fc42 fedora 9.9 MiB filesystem x86_64 3.18-29.fc42 fedora 106.0 B fonts-srpm-macros noarch 1:2.0.5-17.fc41 fedora 55.8 KiB forge-srpm-macros noarch 0.4.0-1.fc42 fedora 38.9 KiB fpc-srpm-macros noarch 1.3-13.fc41 fedora 144.0 B gdb-minimal x86_64 15.2-4.fc42 fedora 12.7 MiB gdbm x86_64 1:1.23-7.fc41 fedora 460.9 KiB gdbm-libs x86_64 1:1.23-7.fc41 fedora 121.9 KiB ghc-srpm-macros noarch 1.9.2-1.fc42 fedora 779.0 B glibc x86_64 2.40.9000-26.fc42 fedora 6.7 MiB glibc-common x86_64 2.40.9000-26.fc42 fedora 1.0 MiB glibc-gconv-extra x86_64 2.40.9000-26.fc42 fedora 8.0 MiB gmp x86_64 1:6.3.0-2.fc41 fedora 811.4 KiB gnat-srpm-macros noarch 6-6.fc41 fedora 1.0 KiB go-srpm-macros noarch 3.6.0-5.fc42 fedora 60.8 KiB jansson x86_64 2.14-1.fc42 fedora 93.1 KiB json-c x86_64 0.18-1.fc42 fedora 83.3 KiB kernel-srpm-macros noarch 1.0-24.fc41 fedora 1.9 KiB keyutils-libs x86_64 1.6.3-4.fc41 fedora 54.4 KiB krb5-libs x86_64 1.21.3-3.fc42 fedora 2.3 MiB libacl x86_64 2.3.2-2.fc41 fedora 40.0 KiB libarchive x86_64 3.7.7-1.fc42 fedora 932.3 KiB libattr x86_64 2.5.2-4.fc41 fedora 28.5 KiB libblkid x86_64 2.40.2-8.fc42 fedora 262.5 KiB libbrotli x86_64 1.1.0-5.fc41 fedora 837.6 KiB libcap x86_64 2.71-1.fc42 fedora 210.8 KiB libcap-ng x86_64 0.8.5-3.fc41 fedora 69.2 KiB libcom_err x86_64 1.47.1-6.fc42 fedora 67.2 KiB libcurl x86_64 8.11.1-2.fc42 fedora 817.3 KiB libeconf x86_64 0.7.5-1.fc42 fedora 66.7 KiB libevent x86_64 2.1.12-14.fc41 fedora 895.7 KiB libfdisk x86_64 2.40.2-8.fc42 fedora 362.9 KiB libffi x86_64 3.4.6-3.fc42 fedora 86.4 KiB libgcc x86_64 14.2.1-6.fc42 copr_base 270.9 KiB libgomp x86_64 14.2.1-6.fc42 copr_base 514.2 KiB libidn2 x86_64 2.3.7-2.fc41 fedora 329.1 KiB libmount x86_64 2.40.2-8.fc42 fedora 355.8 KiB libnghttp2 x86_64 1.64.0-1.fc42 fedora 174.5 KiB libnsl2 x86_64 2.0.1-2.fc41 fedora 57.9 KiB libpkgconf x86_64 2.3.0-1.fc42 fedora 78.2 KiB libpsl x86_64 0.21.5-4.fc41 fedora 80.5 KiB libpwquality x86_64 1.4.5-11.fc41 fedora 417.8 KiB libselinux x86_64 3.8-0.rc3.1.fc42 fedora 191.6 KiB libsemanage x86_64 3.8-0.rc3.1.fc42 fedora 305.3 KiB libsepol x86_64 3.8-0.rc3.1.fc42 fedora 812.3 KiB libsmartcols x86_64 2.40.2-8.fc42 fedora 180.4 KiB libssh x86_64 0.11.1-1.fc42 fedora 569.6 KiB libssh-config noarch 0.11.1-1.fc42 fedora 277.0 B libstdc++ x86_64 14.2.1-6.fc42 copr_base 2.7 MiB libtasn1 x86_64 4.19.0-9.fc41 fedora 175.7 KiB libtirpc x86_64 1.3.6-1.rc3.fc42 fedora 197.6 KiB libtool-ltdl x86_64 2.5.4-1.fc42 fedora 68.1 KiB libunistring x86_64 1.1-8.fc41 fedora 1.7 MiB libuuid x86_64 2.40.2-8.fc42 fedora 41.4 KiB libverto x86_64 0.3.2-9.fc41 fedora 29.5 KiB libxcrypt x86_64 4.4.37-3.fc42 fedora 269.6 KiB libxml2 x86_64 2.12.9-1.fc42 fedora 1.7 MiB libzstd x86_64 1.5.6-2.fc41 fedora 795.9 KiB lua-libs x86_64 5.4.7-1.fc42 fedora 285.0 KiB lua-srpm-macros noarch 1-14.fc41 fedora 1.3 KiB lz4-libs x86_64 1.10.0-1.fc41 fedora 145.5 KiB mpfr x86_64 4.2.1-5.fc41 fedora 832.1 KiB ncurses-base noarch 6.5-2.20240629.fc41 fedora 326.3 KiB ncurses-libs x86_64 6.5-2.20240629.fc41 fedora 975.2 KiB ocaml-srpm-macros noarch 10-3.fc41 fedora 1.9 KiB openblas-srpm-macros noarch 2-18.fc41 fedora 112.0 B openldap x86_64 2.6.8-6.fc42 fedora 647.4 KiB openssl-libs x86_64 1:3.2.2-10.fc42 fedora 7.7 MiB p11-kit x86_64 0.25.5-4.fc42 fedora 2.2 MiB p11-kit-trust x86_64 0.25.5-4.fc42 fedora 403.8 KiB package-notes-srpm-macros noarch 0.5-12.fc41 fedora 1.6 KiB pam x86_64 1.7.0-3.fc42 fedora 1.8 MiB pam-libs x86_64 1.7.0-3.fc42 fedora 139.4 KiB pcre2 x86_64 10.44-1.fc41.1 fedora 653.5 KiB pcre2-syntax noarch 10.44-1.fc41.1 fedora 251.6 KiB perl-srpm-macros noarch 1-56.fc41 fedora 861.0 B pkgconf x86_64 2.3.0-1.fc42 fedora 88.6 KiB pkgconf-m4 noarch 2.3.0-1.fc42 fedora 14.4 KiB pkgconf-pkg-config x86_64 2.3.0-1.fc42 fedora 989.0 B popt x86_64 1.19-7.fc41 fedora 136.9 KiB publicsuffix-list-dafsa noarch 20240107-4.fc41 fedora 67.5 KiB pyproject-srpm-macros noarch 1.16.3-1.fc42 fedora 1.9 KiB python-srpm-macros noarch 3.13-3.fc41 fedora 51.0 KiB qt5-srpm-macros noarch 5.15.15-1.fc42 fedora 500.0 B qt6-srpm-macros noarch 6.8.1-4.fc42 fedora 456.0 B readline x86_64 8.2-11.fc42 fedora 493.1 KiB rpm x86_64 4.20.0-1.fc42 fedora 3.1 MiB rpm-build-libs x86_64 4.20.0-1.fc42 fedora 206.7 KiB rpm-libs x86_64 4.20.0-1.fc42 fedora 726.1 KiB rpm-sequoia x86_64 1.7.0-3.fc42 fedora 2.3 MiB rust-srpm-macros noarch 26.3-3.fc42 fedora 4.8 KiB setup noarch 2.15.0-9.fc42 fedora 720.7 KiB sqlite-libs x86_64 3.47.2-1.fc42 fedora 1.4 MiB systemd-libs x86_64 257.1-1.fc42 fedora 2.2 MiB util-linux-core x86_64 2.40.2-8.fc42 fedora 1.5 MiB xxhash-libs x86_64 0.8.3-1.fc42 fedora 88.5 KiB xz-libs x86_64 1:5.6.3-2.fc42 fedora 218.4 KiB zig-srpm-macros noarch 1-3.fc41 fedora 1.1 KiB zip x86_64 3.0-42.fc42 fedora 695.9 KiB zlib-ng-compat x86_64 2.2.2-1.fc42 fedora 134.0 KiB zstd x86_64 1.5.6-2.fc41 fedora 1.7 MiB Installing groups: Buildsystem building group Transaction Summary: Installing: 154 packages Total size of inbound packages is 52 MiB. Need to download 0 B. After this operation, 178 MiB extra will be used (install 178 MiB, remove 0 B). [1/1] tar-2:1.35-4.fc41.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [1/1] Total 100% | 0.0 B/s | 0.0 B | 00m00s [1/2] bzip2-0:1.0.8-19.fc41.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [2/2] Total 100% | 0.0 B/s | 0.0 B | 00m00s [1/3] redhat-rpm-config-0:296-1.fc42.no 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [3/3] Total 100% | 0.0 B/s | 0.0 B | 00m00s [1/4] rpm-build-0:4.20.0-1.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [4/4] Total 100% | 0.0 B/s | 0.0 B | 00m00s [1/5] unzip-0:6.0-65.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [5/5] Total 100% | 0.0 B/s | 0.0 B | 00m00s [1/6] cpio-0:2.15-2.fc41.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [6/6] Total 100% | 0.0 B/s | 0.0 B | 00m00s [1/7] which-0:2.21-42.fc41.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [7/7] Total 100% | 0.0 B/s | 0.0 B | 00m00s [1/8] bash-0:5.2.37-1.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [8/8] Total 100% | 0.0 B/s | 0.0 B | 00m00s [1/9] coreutils-0:9.5-11.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [9/9] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/10] grep-0:3.11-9.fc41.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [10/10] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/11] patch-0:2.7.6-25.fc41.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [11/11] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/12] sed-0:4.9-3.fc41.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [12/12] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/13] shadow-utils-2:4.17.0-2.fc42.x8 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [13/13] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/14] diffutils-0:3.10-8.fc41.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [14/14] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/15] fedora-release-common-0:42-0.11 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [15/15] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/16] findutils-1:4.10.0-4.fc41.x86_6 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [16/16] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/17] glibc-minimal-langpack-0:2.40.9 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [17/17] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/18] gzip-0:1.13-2.fc41.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [18/18] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/19] info-0:7.1.1-2.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [19/19] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/20] xz-1:5.6.3-2.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [20/20] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/21] util-linux-0:2.40.2-8.fc42.x86_ 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [21/21] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/22] gawk-0:5.3.0-4.fc41.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [22/22] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/23] glibc-0:2.40.9000-26.fc42.x86_6 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [23/23] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/24] libacl-0:2.3.2-2.fc41.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [24/24] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/25] libselinux-0:3.8-0.rc3.1.fc42.x 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [25/25] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/26] bzip2-libs-0:1.0.8-19.fc41.x86_ 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [26/26] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/27] ansible-srpm-macros-0:1-16.fc41 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [27/27] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/28] build-reproducibility-srpm-macr 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [28/28] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/29] dwz-0:0.15-8.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [29/29] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/30] efi-srpm-macros-0:5-13.fc42.noa 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [30/30] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/31] file-0:5.45-8.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [31/31] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/32] fonts-srpm-macros-1:2.0.5-17.fc 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [32/32] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/33] forge-srpm-macros-0:0.4.0-1.fc4 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [33/33] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/34] fpc-srpm-macros-0:1.3-13.fc41.n 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [34/34] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/35] ghc-srpm-macros-0:1.9.2-1.fc42. 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [35/35] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/36] gnat-srpm-macros-0:6-6.fc41.noa 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [36/36] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/37] go-srpm-macros-0:3.6.0-5.fc42.n 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [37/37] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/38] kernel-srpm-macros-0:1.0-24.fc4 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [38/38] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/39] lua-srpm-macros-0:1-14.fc41.noa 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [39/39] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/40] ocaml-srpm-macros-0:10-3.fc41.n 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [40/40] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/41] openblas-srpm-macros-0:2-18.fc4 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [41/41] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/42] package-notes-srpm-macros-0:0.5 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [42/42] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/43] perl-srpm-macros-0:1-56.fc41.no 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [43/43] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/44] pyproject-srpm-macros-0:1.16.3- 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [44/44] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/45] python-srpm-macros-0:3.13-3.fc4 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [45/45] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/46] qt5-srpm-macros-0:5.15.15-1.fc4 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [46/46] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/47] qt6-srpm-macros-0:6.8.1-4.fc42. 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [47/47] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/48] rpm-0:4.20.0-1.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [48/48] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/49] rust-srpm-macros-0:26.3-3.fc42. 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [49/49] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/50] zig-srpm-macros-0:1-3.fc41.noar 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [50/50] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/51] zip-0:3.0-42.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [51/51] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/52] debugedit-0:5.1-2.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [52/52] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/53] elfutils-0:0.192-7.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [53/53] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/54] elfutils-libelf-0:0.192-7.fc42. 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [54/54] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/55] libarchive-0:3.7.7-1.fc42.x86_6 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [55/55] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/56] popt-0:1.19-7.fc41.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [56/56] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/57] readline-0:8.2-11.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [57/57] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/58] rpm-build-libs-0:4.20.0-1.fc42. 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [58/58] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/59] rpm-libs-0:4.20.0-1.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [59/59] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/60] zstd-0:1.5.6-2.fc41.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [60/60] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/61] filesystem-0:3.18-29.fc42.x86_6 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [61/61] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/62] ncurses-libs-0:6.5-2.20240629.f 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [62/62] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/63] coreutils-common-0:9.5-11.fc42. 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [63/63] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/64] gmp-1:6.3.0-2.fc41.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [64/64] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/65] libattr-0:2.5.2-4.fc41.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [65/65] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/66] libcap-0:2.71-1.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [66/66] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/67] openssl-libs-1:3.2.2-10.fc42.x8 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [67/67] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/68] systemd-libs-0:257.1-1.fc42.x86 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [68/68] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/69] pcre2-0:10.44-1.fc41.1.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [69/69] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/70] ed-0:1.20.2-2.fc41.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [70/70] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/71] audit-libs-0:4.0.2-1.fc41.x86_6 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [71/71] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/72] libeconf-0:0.7.5-1.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [72/72] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/73] libsemanage-0:3.8-0.rc3.1.fc42. 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [73/73] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/74] libxcrypt-0:4.4.37-3.fc42.x86_6 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [74/74] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/75] pam-libs-0:1.7.0-3.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [75/75] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/76] setup-0:2.15.0-9.fc42.noarch 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [76/76] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/77] fedora-repos-0:42-0.3.noarch 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [77/77] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/78] glibc-common-0:2.40.9000-26.fc4 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [78/78] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/79] xz-libs-1:5.6.3-2.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [79/79] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/80] libblkid-0:2.40.2-8.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [80/80] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/81] libcap-ng-0:0.8.5-3.fc41.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [81/81] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/82] libfdisk-0:2.40.2-8.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [82/82] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/83] libmount-0:2.40.2-8.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [83/83] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/84] libsmartcols-0:2.40.2-8.fc42.x8 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [84/84] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/85] libuuid-0:2.40.2-8.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [85/85] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/86] util-linux-core-0:2.40.2-8.fc42 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [86/86] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/87] zlib-ng-compat-0:2.2.2-1.fc42.x 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [87/87] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/88] mpfr-0:4.2.1-5.fc41.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [88/88] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/89] glibc-gconv-extra-0:2.40.9000-2 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [89/89] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/90] basesystem-0:11-21.fc41.noarch 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [90/90] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/91] libsepol-0:3.8-0.rc3.1.fc42.x86 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [91/91] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/92] add-determinism-0:0.5.0-1.fc42. 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [92/92] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/93] file-libs-0:5.45-8.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [93/93] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/94] curl-0:8.11.1-2.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [94/94] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/95] elfutils-libs-0:0.192-7.fc42.x8 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [95/95] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/96] elfutils-debuginfod-client-0:0. 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [96/96] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/97] libzstd-0:1.5.6-2.fc41.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [97/97] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/98] libxml2-0:2.12.9-1.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [98/98] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/99] lz4-libs-0:1.10.0-1.fc41.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [99/99] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/100] lua-libs-0:5.4.7-1.fc42.x86_6 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [100/100] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/101] rpm-sequoia-0:1.7.0-3.fc42.x8 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [101/101] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/102] sqlite-libs-0:3.47.2-1.fc42.x 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [102/102] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/103] ncurses-base-0:6.5-2.20240629 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [103/103] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/104] ca-certificates-0:2024.2.69_v 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [104/104] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/105] crypto-policies-0:20241128-1. 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [105/105] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/106] pcre2-syntax-0:10.44-1.fc41.1 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [106/106] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/107] fedora-gpg-keys-0:42-0.3.noar 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [107/107] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/108] fedora-repos-rawhide-0:42-0.3 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [108/108] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/109] elfutils-default-yama-scope-0 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [109/109] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/110] json-c-0:0.18-1.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [110/110] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/111] libgcc-0:14.2.1-6.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [111/111] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/112] libstdc++-0:14.2.1-6.fc42.x86 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [112/112] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/113] authselect-libs-0:1.5.0-8.fc4 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [113/113] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/114] pam-0:1.7.0-3.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [114/114] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/115] authselect-0:1.5.0-8.fc42.x86 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [115/115] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/116] gdbm-libs-1:1.23-7.fc41.x86_6 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [116/116] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/117] libnsl2-0:2.0.1-2.fc41.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [117/117] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/118] libpwquality-0:1.4.5-11.fc41. 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [118/118] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/119] libtirpc-0:1.3.6-1.rc3.fc42.x 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [119/119] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/120] cracklib-0:2.9.11-6.fc41.x86_ 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [120/120] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/121] krb5-libs-0:1.21.3-3.fc42.x86 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [121/121] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/122] libcom_err-0:1.47.1-6.fc42.x8 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [122/122] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/123] keyutils-libs-0:1.6.3-4.fc41. 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [123/123] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/124] libverto-0:0.3.2-9.fc41.x86_6 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [124/124] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/125] libgomp-0:14.2.1-6.fc42.x86_6 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [125/125] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/126] binutils-0:2.43.50-9.fc42.x86 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [126/126] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/127] alternatives-0:1.31-1.fc42.x8 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [127/127] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/128] jansson-0:2.14-1.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [128/128] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/129] pkgconf-pkg-config-0:2.3.0-1. 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [129/129] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/130] pkgconf-0:2.3.0-1.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [130/130] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/131] pkgconf-m4-0:2.3.0-1.fc42.noa 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [131/131] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/132] libpkgconf-0:2.3.0-1.fc42.x86 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [132/132] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/133] gdbm-1:1.23-7.fc41.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [133/133] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/134] libffi-0:3.4.6-3.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [134/134] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/135] p11-kit-0:0.25.5-4.fc42.x86_6 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [135/135] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/136] libtasn1-0:4.19.0-9.fc41.x86_ 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [136/136] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/137] p11-kit-trust-0:0.25.5-4.fc42 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [137/137] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/138] fedora-release-0:42-0.11.noar 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [138/138] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/139] gdb-minimal-0:15.2-4.fc42.x86 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [139/139] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/140] xxhash-libs-0:0.8.3-1.fc42.x8 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [140/140] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/141] fedora-release-identity-basic 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [141/141] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/142] libcurl-0:8.11.1-2.fc42.x86_6 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [142/142] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/143] libbrotli-0:1.1.0-5.fc41.x86_ 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [143/143] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/144] libidn2-0:2.3.7-2.fc41.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [144/144] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/145] libnghttp2-0:1.64.0-1.fc42.x8 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [145/145] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/146] libpsl-0:0.21.5-4.fc41.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [146/146] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/147] libssh-0:0.11.1-1.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [147/147] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/148] openldap-0:2.6.8-6.fc42.x86_6 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [148/148] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/149] libunistring-0:1.1-8.fc41.x86 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [149/149] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/150] publicsuffix-list-dafsa-0:202 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [150/150] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/151] libssh-config-0:0.11.1-1.fc42 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [151/151] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/152] cyrus-sasl-lib-0:2.1.28-27.fc 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [152/152] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/153] libevent-0:2.1.12-14.fc41.x86 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [153/153] Total 100% | 0.0 B/s | 0.0 B | 00m00s [ 1/154] libtool-ltdl-0:2.5.4-1.fc42.x 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded -------------------------------------------------------------------------------- [154/154] Total 100% | 0.0 B/s | 0.0 B | 00m00s Running transaction Importing OpenPGP key 0x105EF944: UserID : "Fedora (42) " Fingerprint: B0F4950458F69E1150C6C5EDC8AC4916105EF944 From : file:///usr/share/distribution-gpg-keys/fedora/RPM-GPG-KEY-fedora-42-primary The key was successfully imported. Importing OpenPGP key 0x105EF944: UserID : "Fedora (42) " Fingerprint: B0F4950458F69E1150C6C5EDC8AC4916105EF944 From : file:///usr/share/distribution-gpg-keys/fedora/RPM-GPG-KEY-fedora-42-primary The key was successfully imported. Importing OpenPGP key 0xE99D6AD1: UserID : "Fedora (41) " Fingerprint: 466CF2D8B60BC3057AA9453ED0622462E99D6AD1 From : file:///usr/share/distribution-gpg-keys/fedora/RPM-GPG-KEY-fedora-41-primary The key was successfully imported. Importing OpenPGP key 0x31645531: UserID : "Fedora (43) " Fingerprint: C6E7F081CF80E13146676E88829B606631645531 From : file:///usr/share/distribution-gpg-keys/fedora/RPM-GPG-KEY-fedora-43-primary The key was successfully imported. [ 1/156] Verify package files 100% | 770.0 B/s | 154.0 B | 00m00s >>> Running pre-transaction scriptlet: filesystem-0:3.18-29.fc42.x86_64 >>> Finished pre-transaction scriptlet: filesystem-0:3.18-29.fc42.x86_64 >>> [RPM] /var/lib/mock/fedora-rawhide-x86_64-1736126189.030111/root/var/cache/d [ 2/156] Prepare transaction 100% | 1.8 KiB/s | 154.0 B | 00m00s [ 3/156] Installing libgcc-0:14.2.1-6. 100% | 133.1 MiB/s | 272.6 KiB | 00m00s [ 4/156] Installing libssh-config-0:0. 100% | 0.0 B/s | 816.0 B | 00m00s [ 5/156] Installing publicsuffix-list- 100% | 66.7 MiB/s | 68.3 KiB | 00m00s [ 6/156] Installing fedora-release-ide 100% | 953.1 KiB/s | 976.0 B | 00m00s [ 7/156] Installing fedora-gpg-keys-0: 100% | 18.7 MiB/s | 172.2 KiB | 00m00s [ 8/156] Installing fedora-repos-rawhi 100% | 0.0 B/s | 2.4 KiB | 00m00s [ 9/156] Installing fedora-repos-0:42- 100% | 0.0 B/s | 5.7 KiB | 00m00s [ 10/156] Installing fedora-release-com 100% | 11.8 MiB/s | 24.1 KiB | 00m00s [ 11/156] Installing fedora-release-0:4 100% | 0.0 B/s | 124.0 B | 00m00s [ 12/156] Installing setup-0:2.15.0-9.f 100% | 47.3 MiB/s | 726.5 KiB | 00m00s >>> [RPM] /etc/hosts created as /etc/hosts.rpmnew [ 13/156] Installing filesystem-0:3.18- 100% | 1.6 MiB/s | 212.6 KiB | 00m00s [ 14/156] Installing basesystem-0:11-21 100% | 0.0 B/s | 124.0 B | 00m00s [ 15/156] Installing pkgconf-m4-0:2.3.0 100% | 0.0 B/s | 14.8 KiB | 00m00s [ 16/156] Installing pcre2-syntax-0:10. 100% | 124.1 MiB/s | 254.1 KiB | 00m00s [ 17/156] Installing ncurses-base-0:6.5 100% | 38.2 MiB/s | 351.7 KiB | 00m00s [ 18/156] Installing glibc-minimal-lang 100% | 0.0 B/s | 124.0 B | 00m00s [ 19/156] Installing ncurses-libs-0:6.5 100% | 137.0 MiB/s | 981.8 KiB | 00m00s [ 20/156] Installing glibc-0:2.40.9000- 100% | 202.7 MiB/s | 6.7 MiB | 00m00s [ 21/156] Installing bash-0:5.2.37-1.fc 100% | 272.3 MiB/s | 8.2 MiB | 00m00s [ 22/156] Installing glibc-common-0:2.4 100% | 130.2 MiB/s | 1.0 MiB | 00m00s [ 23/156] Installing glibc-gconv-extra- 100% | 161.5 MiB/s | 8.1 MiB | 00m00s [ 24/156] Installing zlib-ng-compat-0:2 100% | 131.7 MiB/s | 134.9 KiB | 00m00s [ 25/156] Installing bzip2-libs-0:1.0.8 100% | 79.9 MiB/s | 81.8 KiB | 00m00s [ 26/156] Installing xz-libs-1:5.6.3-2. 100% | 214.3 MiB/s | 219.5 KiB | 00m00s [ 27/156] Installing popt-0:1.19-7.fc41 100% | 35.0 MiB/s | 143.5 KiB | 00m00s [ 28/156] Installing readline-0:8.2-11. 100% | 241.8 MiB/s | 495.3 KiB | 00m00s [ 29/156] Installing libuuid-0:2.40.2-8 100% | 41.5 MiB/s | 42.5 KiB | 00m00s [ 30/156] Installing libblkid-0:2.40.2- 100% | 128.7 MiB/s | 263.6 KiB | 00m00s [ 31/156] Installing gmp-1:6.3.0-2.fc41 100% | 264.9 MiB/s | 813.7 KiB | 00m00s [ 32/156] Installing libattr-0:2.5.2-4. 100% | 28.8 MiB/s | 29.5 KiB | 00m00s [ 33/156] Installing libacl-0:2.3.2-2.f 100% | 39.8 MiB/s | 40.7 KiB | 00m00s [ 34/156] Installing libxcrypt-0:4.4.37 100% | 133.0 MiB/s | 272.3 KiB | 00m00s [ 35/156] Installing libzstd-0:1.5.6-2. 100% | 259.5 MiB/s | 797.2 KiB | 00m00s [ 36/156] Installing elfutils-libelf-0: 100% | 291.7 MiB/s | 1.2 MiB | 00m00s [ 37/156] Installing libstdc++-0:14.2.1 100% | 269.2 MiB/s | 2.7 MiB | 00m00s [ 38/156] Installing libeconf-0:0.7.5-1 100% | 66.8 MiB/s | 68.4 KiB | 00m00s [ 39/156] Installing gdbm-libs-1:1.23-7 100% | 120.7 MiB/s | 123.6 KiB | 00m00s [ 40/156] Installing dwz-0:0.15-8.fc42. 100% | 146.8 MiB/s | 300.6 KiB | 00m00s [ 41/156] Installing mpfr-0:4.2.1-5.fc4 100% | 203.5 MiB/s | 833.7 KiB | 00m00s [ 42/156] Installing gawk-0:5.3.0-4.fc4 100% | 173.2 MiB/s | 1.7 MiB | 00m00s [ 43/156] Installing unzip-0:6.0-65.fc4 100% | 130.8 MiB/s | 401.7 KiB | 00m00s [ 44/156] Installing file-libs-0:5.45-8 100% | 523.1 MiB/s | 9.9 MiB | 00m00s [ 45/156] Installing file-0:5.45-8.fc42 100% | 8.6 MiB/s | 105.2 KiB | 00m00s [ 46/156] Installing crypto-policies-0: 100% | 14.5 MiB/s | 163.7 KiB | 00m00s [ 47/156] Installing pcre2-0:10.44-1.fc 100% | 213.2 MiB/s | 654.9 KiB | 00m00s [ 48/156] Installing grep-0:3.11-9.fc41 100% | 125.4 MiB/s | 1.0 MiB | 00m00s [ 49/156] Installing xz-1:5.6.3-2.fc42. 100% | 123.7 MiB/s | 1.2 MiB | 00m00s [ 50/156] Installing libcap-ng-0:0.8.5- 100% | 69.4 MiB/s | 71.0 KiB | 00m00s [ 51/156] Installing audit-libs-0:4.0.2 100% | 162.8 MiB/s | 333.4 KiB | 00m00s [ 52/156] Installing pam-libs-0:1.7.0-3 100% | 69.2 MiB/s | 141.8 KiB | 00m00s [ 53/156] Installing libcap-0:2.71-1.fc 100% | 70.3 MiB/s | 215.8 KiB | 00m00s [ 54/156] Installing systemd-libs-0:257 100% | 246.7 MiB/s | 2.2 MiB | 00m00s [ 55/156] Installing libsmartcols-0:2.4 100% | 177.1 MiB/s | 181.4 KiB | 00m00s [ 56/156] Installing libsepol-0:3.8-0.r 100% | 264.7 MiB/s | 813.3 KiB | 00m00s [ 57/156] Installing libselinux-0:3.8-0 100% | 94.2 MiB/s | 192.9 KiB | 00m00s [ 58/156] Installing sed-0:4.9-3.fc41.x 100% | 121.3 MiB/s | 869.7 KiB | 00m00s [ 59/156] Installing findutils-1:4.10.0 100% | 154.8 MiB/s | 1.9 MiB | 00m00s [ 60/156] Installing libmount-0:2.40.2- 100% | 116.2 MiB/s | 356.9 KiB | 00m00s [ 61/156] Installing lz4-libs-0:1.10.0- 100% | 143.1 MiB/s | 146.6 KiB | 00m00s [ 62/156] Installing lua-libs-0:5.4.7-1 100% | 139.7 MiB/s | 286.2 KiB | 00m00s [ 63/156] Installing libcom_err-0:1.47. 100% | 66.7 MiB/s | 68.3 KiB | 00m00s [ 64/156] Installing alternatives-0:1.3 100% | 64.8 MiB/s | 66.4 KiB | 00m00s [ 65/156] Installing libffi-0:3.4.6-3.f 100% | 85.7 MiB/s | 87.8 KiB | 00m00s [ 66/156] Installing libtasn1-0:4.19.0- 100% | 86.7 MiB/s | 177.5 KiB | 00m00s [ 67/156] Installing p11-kit-0:0.25.5-4 100% | 158.3 MiB/s | 2.2 MiB | 00m00s [ 68/156] Installing libunistring-0:1.1 100% | 288.5 MiB/s | 1.7 MiB | 00m00s [ 69/156] Installing libidn2-0:2.3.7-2. 100% | 81.8 MiB/s | 335.1 KiB | 00m00s [ 70/156] Installing libpsl-0:0.21.5-4. 100% | 79.7 MiB/s | 81.7 KiB | 00m00s [ 71/156] Installing p11-kit-trust-0:0. 100% | 30.5 MiB/s | 405.5 KiB | 00m00s [ 72/156] Installing zstd-0:1.5.6-2.fc4 100% | 241.6 MiB/s | 1.7 MiB | 00m00s [ 73/156] Installing util-linux-core-0: 100% | 138.9 MiB/s | 1.5 MiB | 00m00s [ 74/156] Installing tar-2:1.35-4.fc41. 100% | 227.6 MiB/s | 3.0 MiB | 00m00s [ 75/156] Installing libsemanage-0:3.8- 100% | 100.0 MiB/s | 307.1 KiB | 00m00s [ 76/156] Installing shadow-utils-2:4.1 100% | 122.7 MiB/s | 4.0 MiB | 00m00s [ 77/156] Installing zip-0:3.0-42.fc42. 100% | 170.9 MiB/s | 699.8 KiB | 00m00s [ 78/156] Installing gdbm-1:1.23-7.fc41 100% | 113.7 MiB/s | 465.8 KiB | 00m00s [ 79/156] Installing cyrus-sasl-lib-0:2 100% | 256.2 MiB/s | 2.3 MiB | 00m00s [ 80/156] Installing libfdisk-0:2.40.2- 100% | 177.7 MiB/s | 364.0 KiB | 00m00s [ 81/156] Installing libxml2-0:2.12.9-1 100% | 240.9 MiB/s | 1.7 MiB | 00m00s [ 82/156] Installing bzip2-0:1.0.8-19.f 100% | 48.9 MiB/s | 100.2 KiB | 00m00s [ 83/156] Installing add-determinism-0: 100% | 296.1 MiB/s | 2.4 MiB | 00m00s [ 84/156] Installing build-reproducibil 100% | 0.0 B/s | 1.0 KiB | 00m00s [ 85/156] Installing sqlite-libs-0:3.47 100% | 286.8 MiB/s | 1.4 MiB | 00m00s [ 86/156] Installing ed-0:1.20.2-2.fc41 100% | 72.8 MiB/s | 149.2 KiB | 00m00s [ 87/156] Installing patch-0:2.7.6-25.f 100% | 131.0 MiB/s | 268.2 KiB | 00m00s [ 88/156] Installing elfutils-default-y 100% | 204.3 KiB/s | 2.0 KiB | 00m00s [ 89/156] Installing elfutils-libs-0:0. 100% | 162.3 MiB/s | 664.7 KiB | 00m00s [ 90/156] Installing cpio-0:2.15-2.fc41 100% | 157.1 MiB/s | 1.1 MiB | 00m00s [ 91/156] Installing diffutils-0:3.10-8 100% | 176.7 MiB/s | 1.6 MiB | 00m00s [ 92/156] Installing json-c-0:0.18-1.fc 100% | 82.6 MiB/s | 84.6 KiB | 00m00s [ 93/156] Installing keyutils-libs-0:1. 100% | 54.5 MiB/s | 55.8 KiB | 00m00s [ 94/156] Installing libverto-0:0.3.2-9 100% | 30.5 MiB/s | 31.3 KiB | 00m00s [ 95/156] Installing libgomp-0:14.2.1-6 100% | 251.7 MiB/s | 515.6 KiB | 00m00s [ 96/156] Installing jansson-0:2.14-1.f 100% | 92.3 MiB/s | 94.5 KiB | 00m00s [ 97/156] Installing libpkgconf-0:2.3.0 100% | 77.5 MiB/s | 79.3 KiB | 00m00s [ 98/156] Installing pkgconf-0:2.3.0-1. 100% | 44.5 MiB/s | 91.1 KiB | 00m00s [ 99/156] Installing pkgconf-pkg-config 100% | 1.7 MiB/s | 1.8 KiB | 00m00s [100/156] Installing xxhash-libs-0:0.8. 100% | 87.8 MiB/s | 89.9 KiB | 00m00s [101/156] Installing libbrotli-0:1.1.0- 100% | 205.0 MiB/s | 839.9 KiB | 00m00s [102/156] Installing libnghttp2-0:1.64. 100% | 171.5 MiB/s | 175.6 KiB | 00m00s [103/156] Installing libtool-ltdl-0:2.5 100% | 33.8 MiB/s | 69.2 KiB | 00m00s [104/156] Installing coreutils-common-0 100% | 254.3 MiB/s | 11.2 MiB | 00m00s [105/156] Installing openssl-libs-1:3.2 100% | 320.4 MiB/s | 7.7 MiB | 00m00s [106/156] Installing coreutils-0:9.5-11 100% | 146.0 MiB/s | 5.4 MiB | 00m00s [107/156] Installing ca-certificates-0: 100% | 1.2 MiB/s | 2.4 MiB | 00m02s [108/156] Installing krb5-libs-0:1.21.3 100% | 177.3 MiB/s | 2.3 MiB | 00m00s [109/156] Installing libarchive-0:3.7.7 100% | 182.5 MiB/s | 934.2 KiB | 00m00s [110/156] Installing libtirpc-0:1.3.6-1 100% | 64.9 MiB/s | 199.4 KiB | 00m00s [111/156] Installing gzip-0:1.13-2.fc41 100% | 77.1 MiB/s | 394.6 KiB | 00m00s [112/156] Installing authselect-libs-0: 100% | 90.8 MiB/s | 837.2 KiB | 00m00s [113/156] Installing cracklib-0:2.9.11- 100% | 34.9 MiB/s | 250.3 KiB | 00m00s [114/156] Installing libpwquality-0:1.4 100% | 52.5 MiB/s | 430.1 KiB | 00m00s [115/156] Installing libnsl2-0:2.0.1-2. 100% | 28.8 MiB/s | 59.1 KiB | 00m00s [116/156] Installing pam-0:1.7.0-3.fc42 100% | 78.9 MiB/s | 1.9 MiB | 00m00s [117/156] Installing libssh-0:0.11.1-1. 100% | 186.1 MiB/s | 571.7 KiB | 00m00s [118/156] Installing rpm-sequoia-0:1.7. 100% | 258.2 MiB/s | 2.3 MiB | 00m00s [119/156] Installing rpm-libs-0:4.20.0- 100% | 236.9 MiB/s | 727.7 KiB | 00m00s [120/156] Installing rpm-build-libs-0:4 100% | 101.3 MiB/s | 207.5 KiB | 00m00s [121/156] Installing libevent-0:2.1.12- 100% | 219.6 MiB/s | 899.5 KiB | 00m00s [122/156] Installing openldap-0:2.6.8-6 100% | 159.0 MiB/s | 651.2 KiB | 00m00s [123/156] Installing libcurl-0:8.11.1-2 100% | 199.8 MiB/s | 818.4 KiB | 00m00s [124/156] Installing elfutils-debuginfo 100% | 27.3 MiB/s | 83.8 KiB | 00m00s [125/156] Installing elfutils-0:0.192-7 100% | 240.4 MiB/s | 2.6 MiB | 00m00s [126/156] Installing binutils-0:2.43.50 100% | 277.5 MiB/s | 25.8 MiB | 00m00s [127/156] Installing gdb-minimal-0:15.2 100% | 310.4 MiB/s | 12.7 MiB | 00m00s [128/156] Installing debugedit-0:5.1-2. 100% | 99.1 MiB/s | 203.0 KiB | 00m00s [129/156] Installing curl-0:8.11.1-2.fc 100% | 24.7 MiB/s | 454.5 KiB | 00m00s [130/156] Installing rpm-0:4.20.0-1.fc4 100% | 89.5 MiB/s | 2.5 MiB | 00m00s [131/156] Installing efi-srpm-macros-0: 100% | 40.2 MiB/s | 41.2 KiB | 00m00s [132/156] Installing lua-srpm-macros-0: 100% | 0.0 B/s | 1.9 KiB | 00m00s [133/156] Installing zig-srpm-macros-0: 100% | 0.0 B/s | 1.7 KiB | 00m00s [134/156] Installing rust-srpm-macros-0 100% | 0.0 B/s | 5.6 KiB | 00m00s [135/156] Installing qt6-srpm-macros-0: 100% | 0.0 B/s | 732.0 B | 00m00s [136/156] Installing qt5-srpm-macros-0: 100% | 0.0 B/s | 776.0 B | 00m00s [137/156] Installing perl-srpm-macros-0 100% | 0.0 B/s | 1.1 KiB | 00m00s [138/156] Installing package-notes-srpm 100% | 0.0 B/s | 2.0 KiB | 00m00s [139/156] Installing openblas-srpm-macr 100% | 0.0 B/s | 392.0 B | 00m00s [140/156] Installing ocaml-srpm-macros- 100% | 0.0 B/s | 2.2 KiB | 00m00s [141/156] Installing kernel-srpm-macros 100% | 0.0 B/s | 2.3 KiB | 00m00s [142/156] Installing gnat-srpm-macros-0 100% | 0.0 B/s | 1.3 KiB | 00m00s [143/156] Installing ghc-srpm-macros-0: 100% | 0.0 B/s | 1.0 KiB | 00m00s [144/156] Installing fpc-srpm-macros-0: 100% | 0.0 B/s | 420.0 B | 00m00s [145/156] Installing ansible-srpm-macro 100% | 0.0 B/s | 36.2 KiB | 00m00s [146/156] Installing fonts-srpm-macros- 100% | 55.7 MiB/s | 57.0 KiB | 00m00s [147/156] Installing forge-srpm-macros- 100% | 39.3 MiB/s | 40.3 KiB | 00m00s [148/156] Installing go-srpm-macros-0:3 100% | 60.5 MiB/s | 62.0 KiB | 00m00s [149/156] Installing python-srpm-macros 100% | 50.9 MiB/s | 52.2 KiB | 00m00s [150/156] Installing redhat-rpm-config- 100% | 62.9 MiB/s | 193.2 KiB | 00m00s [151/156] Installing rpm-build-0:4.20.0 100% | 49.5 MiB/s | 202.9 KiB | 00m00s [152/156] Installing pyproject-srpm-mac 100% | 1.2 MiB/s | 2.5 KiB | 00m00s [153/156] Installing util-linux-0:2.40. 100% | 92.9 MiB/s | 3.7 MiB | 00m00s [154/156] Installing authselect-0:1.5.0 100% | 39.5 MiB/s | 161.9 KiB | 00m00s [155/156] Installing which-0:2.21-42.fc 100% | 40.2 MiB/s | 82.4 KiB | 00m00s [156/156] Installing info-0:7.1.1-2.fc4 100% | 137.4 KiB/s | 362.2 KiB | 00m03s Warning: skipped OpenPGP checks for 3 packages from repository: copr_base Complete! Finish: installing minimal buildroot with dnf5 Start: creating root cache Finish: creating root cache Finish: chroot init INFO: Installed packages: INFO: add-determinism-0.5.0-1.fc42.x86_64 alternatives-1.31-1.fc42.x86_64 ansible-srpm-macros-1-16.fc41.noarch audit-libs-4.0.2-1.fc41.x86_64 authselect-1.5.0-8.fc42.x86_64 authselect-libs-1.5.0-8.fc42.x86_64 basesystem-11-21.fc41.noarch bash-5.2.37-1.fc42.x86_64 binutils-2.43.50-9.fc42.x86_64 build-reproducibility-srpm-macros-0.5.0-1.fc42.noarch bzip2-1.0.8-19.fc41.x86_64 bzip2-libs-1.0.8-19.fc41.x86_64 ca-certificates-2024.2.69_v8.0.401-4.fc42.noarch coreutils-9.5-11.fc42.x86_64 coreutils-common-9.5-11.fc42.x86_64 cpio-2.15-2.fc41.x86_64 cracklib-2.9.11-6.fc41.x86_64 crypto-policies-20241128-1.gitbb7b0b0.fc42.noarch curl-8.11.1-2.fc42.x86_64 cyrus-sasl-lib-2.1.28-27.fc41.x86_64 debugedit-5.1-2.fc42.x86_64 diffutils-3.10-8.fc41.x86_64 dwz-0.15-8.fc42.x86_64 ed-1.20.2-2.fc41.x86_64 efi-srpm-macros-5-13.fc42.noarch elfutils-0.192-7.fc42.x86_64 elfutils-debuginfod-client-0.192-7.fc42.x86_64 elfutils-default-yama-scope-0.192-7.fc42.noarch elfutils-libelf-0.192-7.fc42.x86_64 elfutils-libs-0.192-7.fc42.x86_64 fedora-gpg-keys-42-0.3.noarch fedora-release-42-0.11.noarch fedora-release-common-42-0.11.noarch fedora-release-identity-basic-42-0.11.noarch fedora-repos-42-0.3.noarch fedora-repos-rawhide-42-0.3.noarch file-5.45-8.fc42.x86_64 file-libs-5.45-8.fc42.x86_64 filesystem-3.18-29.fc42.x86_64 findutils-4.10.0-4.fc41.x86_64 fonts-srpm-macros-2.0.5-17.fc41.noarch forge-srpm-macros-0.4.0-1.fc42.noarch fpc-srpm-macros-1.3-13.fc41.noarch gawk-5.3.0-4.fc41.x86_64 gdb-minimal-15.2-4.fc42.x86_64 gdbm-1.23-7.fc41.x86_64 gdbm-libs-1.23-7.fc41.x86_64 ghc-srpm-macros-1.9.2-1.fc42.noarch glibc-2.40.9000-26.fc42.x86_64 glibc-common-2.40.9000-26.fc42.x86_64 glibc-gconv-extra-2.40.9000-26.fc42.x86_64 glibc-minimal-langpack-2.40.9000-26.fc42.x86_64 gmp-6.3.0-2.fc41.x86_64 gnat-srpm-macros-6-6.fc41.noarch go-srpm-macros-3.6.0-5.fc42.noarch gpg-pubkey-105ef944-65ca83d1 gpg-pubkey-31645531-66b6dccf gpg-pubkey-e99d6ad1-64d2612c grep-3.11-9.fc41.x86_64 gzip-1.13-2.fc41.x86_64 info-7.1.1-2.fc42.x86_64 jansson-2.14-1.fc42.x86_64 json-c-0.18-1.fc42.x86_64 kernel-srpm-macros-1.0-24.fc41.noarch keyutils-libs-1.6.3-4.fc41.x86_64 krb5-libs-1.21.3-3.fc42.x86_64 libacl-2.3.2-2.fc41.x86_64 libarchive-3.7.7-1.fc42.x86_64 libattr-2.5.2-4.fc41.x86_64 libblkid-2.40.2-8.fc42.x86_64 libbrotli-1.1.0-5.fc41.x86_64 libcap-2.71-1.fc42.x86_64 libcap-ng-0.8.5-3.fc41.x86_64 libcom_err-1.47.1-6.fc42.x86_64 libcurl-8.11.1-2.fc42.x86_64 libeconf-0.7.5-1.fc42.x86_64 libevent-2.1.12-14.fc41.x86_64 libfdisk-2.40.2-8.fc42.x86_64 libffi-3.4.6-3.fc42.x86_64 libgcc-14.2.1-6.fc42.x86_64 libgomp-14.2.1-6.fc42.x86_64 libidn2-2.3.7-2.fc41.x86_64 libmount-2.40.2-8.fc42.x86_64 libnghttp2-1.64.0-1.fc42.x86_64 libnsl2-2.0.1-2.fc41.x86_64 libpkgconf-2.3.0-1.fc42.x86_64 libpsl-0.21.5-4.fc41.x86_64 libpwquality-1.4.5-11.fc41.x86_64 libselinux-3.8-0.rc3.1.fc42.x86_64 libsemanage-3.8-0.rc3.1.fc42.x86_64 libsepol-3.8-0.rc3.1.fc42.x86_64 libsmartcols-2.40.2-8.fc42.x86_64 libssh-0.11.1-1.fc42.x86_64 libssh-config-0.11.1-1.fc42.noarch libstdc++-14.2.1-6.fc42.x86_64 libtasn1-4.19.0-9.fc41.x86_64 libtirpc-1.3.6-1.rc3.fc42.x86_64 libtool-ltdl-2.5.4-1.fc42.x86_64 libunistring-1.1-8.fc41.x86_64 libuuid-2.40.2-8.fc42.x86_64 libverto-0.3.2-9.fc41.x86_64 libxcrypt-4.4.37-3.fc42.x86_64 libxml2-2.12.9-1.fc42.x86_64 libzstd-1.5.6-2.fc41.x86_64 lua-libs-5.4.7-1.fc42.x86_64 lua-srpm-macros-1-14.fc41.noarch lz4-libs-1.10.0-1.fc41.x86_64 mpfr-4.2.1-5.fc41.x86_64 ncurses-base-6.5-2.20240629.fc41.noarch ncurses-libs-6.5-2.20240629.fc41.x86_64 ocaml-srpm-macros-10-3.fc41.noarch openblas-srpm-macros-2-18.fc41.noarch openldap-2.6.8-6.fc42.x86_64 openssl-libs-3.2.2-10.fc42.x86_64 p11-kit-0.25.5-4.fc42.x86_64 p11-kit-trust-0.25.5-4.fc42.x86_64 package-notes-srpm-macros-0.5-12.fc41.noarch pam-1.7.0-3.fc42.x86_64 pam-libs-1.7.0-3.fc42.x86_64 patch-2.7.6-25.fc41.x86_64 pcre2-10.44-1.fc41.1.x86_64 pcre2-syntax-10.44-1.fc41.1.noarch perl-srpm-macros-1-56.fc41.noarch pkgconf-2.3.0-1.fc42.x86_64 pkgconf-m4-2.3.0-1.fc42.noarch pkgconf-pkg-config-2.3.0-1.fc42.x86_64 popt-1.19-7.fc41.x86_64 publicsuffix-list-dafsa-20240107-4.fc41.noarch pyproject-srpm-macros-1.16.3-1.fc42.noarch python-srpm-macros-3.13-3.fc41.noarch qt5-srpm-macros-5.15.15-1.fc42.noarch qt6-srpm-macros-6.8.1-4.fc42.noarch readline-8.2-11.fc42.x86_64 redhat-rpm-config-296-1.fc42.noarch rpm-4.20.0-1.fc42.x86_64 rpm-build-4.20.0-1.fc42.x86_64 rpm-build-libs-4.20.0-1.fc42.x86_64 rpm-libs-4.20.0-1.fc42.x86_64 rpm-sequoia-1.7.0-3.fc42.x86_64 rust-srpm-macros-26.3-3.fc42.noarch sed-4.9-3.fc41.x86_64 setup-2.15.0-9.fc42.noarch shadow-utils-4.17.0-2.fc42.x86_64 sqlite-libs-3.47.2-1.fc42.x86_64 systemd-libs-257.1-1.fc42.x86_64 tar-1.35-4.fc41.x86_64 unzip-6.0-65.fc42.x86_64 util-linux-2.40.2-8.fc42.x86_64 util-linux-core-2.40.2-8.fc42.x86_64 which-2.21-42.fc41.x86_64 xxhash-libs-0.8.3-1.fc42.x86_64 xz-5.6.3-2.fc42.x86_64 xz-libs-5.6.3-2.fc42.x86_64 zig-srpm-macros-1-3.fc41.noarch zip-3.0-42.fc42.x86_64 zlib-ng-compat-2.2.2-1.fc42.x86_64 zstd-1.5.6-2.fc41.x86_64 Start: buildsrpm Start: rpmbuild -bs Building target platforms: x86_64 Building for target x86_64 setting SOURCE_DATE_EPOCH=1733184000 Wrote: /builddir/build/SRPMS/llama-cpp-b4094-8.fc42.src.rpm Finish: rpmbuild -bs INFO: chroot_scan: 1 files copied to /var/lib/copr-rpmbuild/results/chroot_scan INFO: /var/lib/mock/fedora-rawhide-x86_64-1736126189.030111/root/var/log/dnf5.log INFO: chroot_scan: creating tarball /var/lib/copr-rpmbuild/results/chroot_scan.tar.gz /bin/tar: Removing leading `/' from member names Finish: buildsrpm INFO: Done(/var/lib/copr-rpmbuild/workspace/workdir-g_cwufk3/llama-cpp/llama-cpp.spec) Config(child) 0 minutes 16 seconds INFO: Results and/or logs in: /var/lib/copr-rpmbuild/results INFO: Cleaning up build root ('cleanup_on_success=True') Start: clean chroot INFO: unmounting tmpfs. Finish: clean chroot INFO: Start(/var/lib/copr-rpmbuild/results/llama-cpp-b4094-8.fc42.src.rpm) Config(fedora-rawhide-x86_64) Start(bootstrap): chroot init INFO: mounting tmpfs at /var/lib/mock/fedora-rawhide-x86_64-bootstrap-1736126189.030111/root. INFO: reusing tmpfs at /var/lib/mock/fedora-rawhide-x86_64-bootstrap-1736126189.030111/root. INFO: calling preinit hooks INFO: enabled root cache INFO: enabled package manager cache Start(bootstrap): cleaning package manager metadata Finish(bootstrap): cleaning package manager metadata Finish(bootstrap): chroot init Start: chroot init INFO: mounting tmpfs at /var/lib/mock/fedora-rawhide-x86_64-1736126189.030111/root. INFO: calling preinit hooks INFO: enabled root cache Start: unpacking root cache Finish: unpacking root cache INFO: enabled package manager cache Start: cleaning package manager metadata Finish: cleaning package manager metadata INFO: enabled HW Info plugin INFO: Buildroot is handled by package management downloaded with a bootstrap image: rpm-4.20.0-1.fc42.x86_64 rpm-sequoia-1.7.0-3.fc42.x86_64 dnf5-5.2.8.1-2.fc42.x86_64 dnf5-plugins-5.2.8.1-2.fc42.x86_64 Finish: chroot init Start: build phase for llama-cpp-b4094-8.fc42.src.rpm Start: build setup for llama-cpp-b4094-8.fc42.src.rpm Building target platforms: x86_64 Building for target x86_64 setting SOURCE_DATE_EPOCH=1733184000 Wrote: /builddir/build/SRPMS/llama-cpp-b4094-8.fc42.src.rpm Updating and loading repositories: fedora 100% | 907.4 KiB/s | 25.4 KiB | 00m00s Copr repository 100% | 38.2 KiB/s | 1.5 KiB | 00m00s Repositories loaded. Package "curl-8.11.1-2.fc42.x86_64" is already installed. Package Arch Version Repository Size Installing: cmake x86_64 3.31.3-1.fc42 fedora 32.9 MiB gcc-c++ x86_64 14.2.1-6.fc42 copr_base 37.7 MiB hipblas-devel x86_64 6.3.0-1.fc42 copr_base 3.1 MiB hipcc-libomp-devel x86_64 18-31.rocm6.3.1.fc42 copr_base 0.0 B langpacks-en noarch 4.2-2.fc42 fedora 400.0 B libcurl-devel x86_64 8.11.1-2.fc42 fedora 1.3 MiB openmpi x86_64 5.0.6-1.fc42 fedora 7.0 MiB pthreadpool-devel x86_64 0.0^git20230829.4fe0e1e-5.fc41 fedora 99.1 KiB rocblas-devel x86_64 6.3.0-2.fc42 copr_base 2.8 MiB rocm-comgr-devel x86_64 18-31.rocm6.3.1.fc42 copr_base 103.0 KiB rocm-hip-devel x86_64 6.3.1-1.fc42 copr_base 2.7 MiB rocm-rpm-macros x86_64 6.3.0-1.fc42 fedora 18.7 KiB rocm-rpm-macros-modules x86_64 6.3.0-1.fc42 fedora 24.3 KiB rocm-runtime-devel x86_64 6.3.1-1.fc42 copr_base 565.6 KiB wget2-wget x86_64 2.2.0-1.fc42 fedora 42.0 B xxd x86_64 2:9.1.984-1.fc42 fedora 43.8 KiB Installing dependencies: abattis-cantarell-vf-fonts noarch 0.301-13.fc41 fedora 192.7 KiB annobin-docs noarch 12.80-1.fc42 fedora 98.6 KiB annobin-plugin-gcc x86_64 12.80-1.fc42 fedora 992.8 KiB brotli x86_64 1.1.0-5.fc41 fedora 31.8 KiB brotli-devel x86_64 1.1.0-5.fc41 fedora 65.6 KiB clang-resource-filesystem x86_64 19.1.6-2.fc42 fedora 15.3 KiB cmake-data noarch 3.31.3-1.fc42 fedora 8.5 MiB cmake-filesystem x86_64 3.31.3-1.fc42 fedora 0.0 B cmake-rpm-macros noarch 3.31.3-1.fc42 fedora 7.5 KiB cpp x86_64 14.2.1-6.fc42 copr_base 34.6 MiB dbus x86_64 1:1.16.0-1.fc42 fedora 0.0 B dbus-broker x86_64 36-4.fc41 fedora 382.8 KiB dbus-common noarch 1:1.16.0-1.fc42 fedora 11.2 KiB default-fonts-core-sans noarch 4.2-2.fc42 fedora 11.9 KiB emacs-filesystem noarch 1:30.0-3.fc41 fedora 0.0 B environment-modules x86_64 5.5.0-1.fc42 fedora 1.8 MiB expat x86_64 2.6.4-1.fc42 fedora 285.5 KiB fonts-filesystem noarch 1:2.0.5-17.fc41 fedora 0.0 B gcc x86_64 14.2.1-6.fc42 copr_base 103.4 MiB gcc-plugin-annobin x86_64 14.2.1-6.fc42 copr_base 59.6 KiB glibc-devel x86_64 2.40.9000-26.fc42 fedora 2.3 MiB gnupg2 x86_64 2.4.5-4.fc42 fedora 9.6 MiB gnutls x86_64 3.8.8-1.fc42 fedora 3.2 MiB gnutls-dane x86_64 3.8.8-1.fc42 fedora 70.0 KiB google-noto-fonts-common noarch 20240901-1.fc42 fedora 17.5 KiB google-noto-sans-mono-vf-fonts noarch 20240901-1.fc42 fedora 561.2 KiB google-noto-sans-vf-fonts noarch 20240901-1.fc42 fedora 1.2 MiB google-noto-serif-vf-fonts noarch 20240901-1.fc42 fedora 1.5 MiB gpgme x86_64 1.24.0-1.fc42 fedora 586.7 KiB groff-base x86_64 1.23.0-7.fc41 fedora 3.8 MiB hipblas x86_64 6.3.0-1.fc42 copr_base 1.1 MiB hipblas-common-devel noarch 6.3.0-1.fc42 copr_base 16.4 KiB hipcc x86_64 18-31.rocm6.3.1.fc42 copr_base 771.0 KiB hiredis x86_64 1.2.0-4.fc42 fedora 114.0 KiB hwdata noarch 0.390-1.fc42 fedora 9.3 MiB hwloc-libs x86_64 2.11.2-1.fc42 fedora 2.9 MiB jsoncpp x86_64 1.9.5-8.fc41 fedora 253.4 KiB kernel-headers x86_64 6.13.0-0.rc5.42.fc42 fedora 6.5 MiB keyutils-libs-devel x86_64 1.6.3-4.fc41 fedora 48.2 KiB krb5-devel x86_64 1.21.3-3.fc42 fedora 705.9 KiB langpacks-core-en noarch 4.2-2.fc42 fedora 398.0 B langpacks-fonts-en noarch 4.2-2.fc42 fedora 341.0 B less x86_64 668-1.fc42 fedora 406.4 KiB libassuan x86_64 2.5.7-2.fc41 fedora 163.8 KiB libb2 x86_64 0.98.1-12.fc41 fedora 42.2 KiB libcbor x86_64 0.11.0-2.fc41 fedora 73.9 KiB libcom_err-devel x86_64 1.47.1-6.fc42 fedora 16.7 KiB libdrm x86_64 2.4.124-1.fc42 fedora 397.7 KiB libedit x86_64 3.1-54.20250104cvs.fc42 fedora 239.3 KiB libfabric x86_64 1.22.0-1.fc41 fedora 5.2 MiB libfido2 x86_64 1.15.0-2.fc41 fedora 238.2 KiB libgcrypt x86_64 1.11.0-4.fc42 fedora 1.5 MiB libgfortran x86_64 14.2.1-6.fc42 copr_base 3.0 MiB libgpg-error x86_64 1.51-1.fc42 fedora 887.2 KiB libibverbs x86_64 54.0-3.fc42 fedora 1.2 MiB libidn2-devel x86_64 2.3.7-2.fc41 fedora 252.1 KiB libkadm5 x86_64 1.21.3-3.fc42 fedora 211.6 KiB libksba x86_64 1.6.7-2.fc41 fedora 398.4 KiB libmpc x86_64 1.3.1-6.fc41 fedora 164.7 KiB libnghttp2-devel x86_64 1.64.0-1.fc42 fedora 295.4 KiB libnl3 x86_64 3.11.0-1.fc42 fedora 1.0 MiB libomp x86_64 19.1.6-2.fc42 fedora 1.8 MiB libomp-devel x86_64 19.1.6-2.fc42 fedora 31.2 MiB libpciaccess x86_64 0.16-13.fc41 fedora 44.6 KiB libpipeline x86_64 1.5.8-1.fc42 fedora 149.1 KiB libpsl-devel x86_64 0.21.5-4.fc41 fedora 110.3 KiB libpsm2 x86_64 12.0.1-1.fc42 fedora 440.0 KiB libquadmath x86_64 14.2.1-6.fc42 copr_base 323.5 KiB librdmacm x86_64 54.0-3.fc42 fedora 155.0 KiB libseccomp x86_64 2.5.5-2.fc41 fedora 173.3 KiB libselinux-devel x86_64 3.8-0.rc3.1.fc42 fedora 126.8 KiB libsepol-devel x86_64 3.8-0.rc3.1.fc42 fedora 120.8 KiB libssh-devel x86_64 0.11.1-1.fc42 fedora 177.8 KiB libstdc++-devel x86_64 14.2.1-6.fc42 copr_base 15.4 MiB libusb1 x86_64 1.0.27-4.fc42 fedora 170.6 KiB libuv x86_64 1:1.49.2-1.fc42 fedora 569.4 KiB libverto-devel x86_64 0.3.2-9.fc41 fedora 25.7 KiB libxcrypt-devel x86_64 4.4.37-3.fc42 fedora 30.5 KiB llvm-libs x86_64 19.1.6-2.fc42 fedora 123.0 MiB logrotate x86_64 3.22.0-2.fc41 fedora 153.1 KiB make x86_64 1:4.4.1-9.fc42 fedora 1.8 MiB man-db x86_64 2.13.0-1.fc42 fedora 2.8 MiB mpdecimal x86_64 2.5.1-16.fc41 fedora 204.9 KiB munge x86_64 0.5.16-3.fc41 fedora 346.3 KiB munge-libs x86_64 0.5.16-3.fc41 fedora 32.1 KiB ncurses x86_64 6.5-2.20240629.fc41 fedora 627.3 KiB nettle x86_64 3.10-3.fc41 fedora 793.0 KiB npth x86_64 1.8-1.fc42 fedora 53.6 KiB numactl-libs x86_64 2.0.19-1.fc42 fedora 54.4 KiB openssh x86_64 9.9p1-5.fc42 fedora 1.4 MiB openssh-clients x86_64 9.9p1-5.fc42 fedora 2.7 MiB openssl-devel x86_64 1:3.2.2-10.fc42 fedora 4.3 MiB orangefs x86_64 2.9.8-12.fc41 fedora 3.1 MiB pcre2-devel x86_64 10.44-1.fc41.1 fedora 2.0 MiB pcre2-utf16 x86_64 10.44-1.fc41.1 fedora 590.1 KiB pcre2-utf32 x86_64 10.44-1.fc41.1 fedora 562.0 KiB perl-AutoLoader noarch 5.74-512.fc42 fedora 20.5 KiB perl-B x86_64 1.89-512.fc42 fedora 498.0 KiB perl-Carp noarch 1.54-511.fc41 fedora 46.6 KiB perl-Class-Struct noarch 0.68-512.fc42 fedora 25.4 KiB perl-Data-Dumper x86_64 2.189-512.fc41 fedora 111.7 KiB perl-Digest noarch 1.20-511.fc41 fedora 35.3 KiB perl-Digest-MD5 x86_64 2.59-5.fc41 fedora 59.8 KiB perl-DynaLoader x86_64 1.56-512.fc42 fedora 32.1 KiB perl-Encode x86_64 4:3.21-511.fc41 fedora 4.7 MiB perl-Errno x86_64 1.38-512.fc42 fedora 8.4 KiB perl-Exporter noarch 5.78-511.fc41 fedora 54.3 KiB perl-Fcntl x86_64 1.18-512.fc42 fedora 49.0 KiB perl-File-Basename noarch 2.86-512.fc42 fedora 14.0 KiB perl-File-Copy noarch 2.41-512.fc42 fedora 19.6 KiB perl-File-Path noarch 2.18-511.fc41 fedora 63.5 KiB perl-File-Temp noarch 1:0.231.100-511.fc41 fedora 162.3 KiB perl-File-Which noarch 1.27-12.fc41 fedora 30.4 KiB perl-File-stat noarch 1.14-512.fc42 fedora 12.5 KiB perl-FileHandle noarch 2.05-512.fc42 fedora 9.3 KiB perl-Getopt-Long noarch 1:2.58-2.fc41 fedora 144.5 KiB perl-Getopt-Std noarch 1.14-512.fc42 fedora 11.2 KiB perl-HTTP-Tiny noarch 0.090-1.fc42 fedora 154.4 KiB perl-IO x86_64 1.55-512.fc42 fedora 151.1 KiB perl-IO-Socket-IP noarch 0.43-1.fc42 fedora 100.3 KiB perl-IO-Socket-SSL noarch 2.089-1.fc42 fedora 703.3 KiB perl-IPC-Open3 noarch 1.22-512.fc42 fedora 22.5 KiB perl-MIME-Base32 noarch 1.303-21.fc41 fedora 30.7 KiB perl-MIME-Base64 x86_64 3.16-511.fc41 fedora 46.1 KiB perl-Net-SSLeay x86_64 1.94-7.fc41 fedora 1.3 MiB perl-POSIX x86_64 2.20-512.fc42 fedora 235.1 KiB perl-PathTools x86_64 3.91-511.fc41 fedora 180.0 KiB perl-Pod-Escapes noarch 1:1.07-511.fc41 fedora 24.9 KiB perl-Pod-Perldoc noarch 3.28.01-512.fc41 fedora 163.7 KiB perl-Pod-Simple noarch 1:3.45-511.fc41 fedora 560.9 KiB perl-Pod-Usage noarch 4:2.03-511.fc41 fedora 84.8 KiB perl-Scalar-List-Utils x86_64 5:1.68-1.fc42 fedora 148.9 KiB perl-SelectSaver noarch 1.02-512.fc42 fedora 2.2 KiB perl-Socket x86_64 4:2.038-511.fc41 fedora 124.0 KiB perl-Storable x86_64 1:3.32-511.fc41 fedora 232.4 KiB perl-Symbol noarch 1.09-512.fc42 fedora 6.8 KiB perl-Term-ANSIColor noarch 5.01-512.fc41 fedora 97.5 KiB perl-Term-Cap noarch 1.18-511.fc41 fedora 29.3 KiB perl-Text-ParseWords noarch 3.31-511.fc41 fedora 13.6 KiB perl-Text-Tabs+Wrap noarch 2024.001-511.fc41 fedora 22.6 KiB perl-Time-Local noarch 2:1.350-511.fc41 fedora 69.0 KiB perl-URI noarch 5.31-1.fc42 fedora 257.0 KiB perl-base noarch 2.27-512.fc42 fedora 12.5 KiB perl-constant noarch 1.33-512.fc41 fedora 26.2 KiB perl-if noarch 0.61.000-512.fc42 fedora 5.8 KiB perl-interpreter x86_64 4:5.40.0-512.fc42 fedora 122.3 KiB perl-libnet noarch 3.15-512.fc41 fedora 289.4 KiB perl-libs x86_64 4:5.40.0-512.fc42 fedora 9.9 MiB perl-locale noarch 1.12-512.fc42 fedora 6.5 KiB perl-mro x86_64 1.29-512.fc42 fedora 45.6 KiB perl-overload noarch 1.37-512.fc42 fedora 71.5 KiB perl-overloading noarch 0.02-512.fc42 fedora 4.8 KiB perl-parent noarch 1:0.244-1.fc42 fedora 10.3 KiB perl-podlators noarch 1:6.0.2-2.fc41 fedora 317.5 KiB perl-vars noarch 1.05-512.fc42 fedora 3.9 KiB pmix x86_64 4.2.8-3.fc41 fedora 2.0 MiB procps-ng x86_64 4.0.4-4.fc41 fedora 1.0 MiB protobuf-c x86_64 1.5.0-4.fc41 fedora 54.0 KiB prrte x86_64 3.0.6-1.fc42 fedora 174.9 KiB prrte-libs x86_64 3.0.6-1.fc42 fedora 1.7 MiB pthreadpool x86_64 0.0^git20230829.4fe0e1e-5.fc41 fedora 113.3 KiB publicsuffix-list noarch 20240107-4.fc41 fedora 318.0 KiB python-pip-wheel noarch 24.3.1-1.fc42 fedora 1.2 MiB python3 x86_64 3.13.1-2.fc42 fedora 30.6 KiB python3-libs x86_64 3.13.1-2.fc42 fedora 39.8 MiB rhash x86_64 1.4.5-1.fc42 fedora 359.3 KiB rocblas x86_64 6.3.0-2.fc42 copr_base 3.7 GiB rocm-clang x86_64 18-31.rocm6.3.1.fc42 copr_base 125.2 MiB rocm-clang-devel x86_64 18-31.rocm6.3.1.fc42 copr_base 22.0 MiB rocm-clang-libs x86_64 18-31.rocm6.3.1.fc42 copr_base 116.3 MiB rocm-clang-runtime-devel x86_64 18-31.rocm6.3.1.fc42 copr_base 6.9 MiB rocm-comgr x86_64 18-31.rocm6.3.1.fc42 copr_base 150.2 MiB rocm-device-libs x86_64 18-31.rocm6.3.1.fc42 copr_base 3.2 MiB rocm-hip x86_64 6.3.1-1.fc42 copr_base 23.2 MiB rocm-lld x86_64 18-31.rocm6.3.1.fc42 copr_base 7.4 MiB rocm-llvm x86_64 18-31.rocm6.3.1.fc42 copr_base 87.5 MiB rocm-llvm-devel x86_64 18-31.rocm6.3.1.fc42 copr_base 24.4 MiB rocm-llvm-libs x86_64 18-31.rocm6.3.1.fc42 copr_base 105.0 MiB rocm-llvm-static x86_64 18-31.rocm6.3.1.fc42 copr_base 241.4 MiB rocm-runtime x86_64 6.3.1-1.fc42 copr_base 2.8 MiB rocsolver x86_64 6.3.0-1.fc42 copr_base 117.8 MiB systemd x86_64 257.1-1.fc42 fedora 17.4 MiB systemd-pam x86_64 257.1-1.fc42 fedora 1.1 MiB systemd-rpm-macros noarch 257.1-1.fc42 fedora 10.7 KiB tcl x86_64 1:8.6.15-6.fc42 fedora 4.2 MiB tcsh x86_64 6.24.14-1.fc42 fedora 1.2 MiB torque-libs x86_64 6.1.3-13.fc42 fedora 458.3 KiB tpm2-tss x86_64 4.1.3-5.fc42 fedora 1.6 MiB tzdata noarch 2024b-1.fc42 fedora 1.6 MiB ucx x86_64 1.17.0-3.fc42 fedora 2.4 MiB unbound-libs x86_64 1.22.0-8.fc42 fedora 1.4 MiB vim-filesystem noarch 2:9.1.984-1.fc42 fedora 40.0 B wget2 x86_64 2.2.0-1.fc42 fedora 1.0 MiB wget2-libs x86_64 2.2.0-1.fc42 fedora 364.7 KiB zlib-ng-compat-devel x86_64 2.2.2-1.fc42 fedora 106.8 KiB Transaction Summary: Installing: 211 packages Total size of inbound packages is 655 MiB. Need to download 321 MiB. After this operation, 5 GiB extra will be used (install 5 GiB, remove 0 B). [1/5] cmake-0:3.31.3-1.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [2/6] gcc-c++-0:14.2.1-6.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 3/12] rocm-comgr-devel-0:18-31.rocm6. 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 4/13] rocm-hip-devel-0:6.3.1-1.fc42.x 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 5/14] rocm-rpm-macros-0:6.3.0-1.fc42. 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 6/16] rocm-runtime-devel-0:6.3.1-1.fc 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 7/19] hwloc-libs-0:2.11.2-1.fc42.x86_ 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 8/28] cmake-data-0:3.31.3-1.fc42.noar 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 9/29] cmake-filesystem-0:3.31.3-1.fc4 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [10/30] expat-0:2.6.4-1.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [11/31] jsoncpp-0:1.9.5-8.fc41.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [12/32] libuv-1:1.49.2-1.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [13/33] make-1:4.4.1-9.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [14/34] rhash-0:1.4.5-1.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [15/35] libmpc-0:1.3.1-6.fc41.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [16/38] perl-File-Basename-0:2.86-512.f 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [17/39] perl-File-Copy-0:2.41-512.fc42. 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [18/40] perl-File-Which-0:1.27-12.fc41. 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [19/41] perl-Getopt-Std-0:1.14-512.fc42 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [20/42] perl-PathTools-0:3.91-511.fc41. 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [21/43] perl-Scalar-List-Utils-5:1.68-1 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [22/44] perl-URI-0:5.31-1.fc42.noarch 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [23/45] perl-interpreter-4:5.40.0-512.f 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [24/46] environment-modules-0:5.5.0-1.f 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [25/47] default-fonts-core-sans-0:4.2-2 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [26/53] numactl-libs-0:2.0.19-1.fc42.x8 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [27/54] libedit-0:3.1-54.20250104cvs.fc 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [28/60] gpgme-0:1.24.0-1.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [29/62] emacs-filesystem-1:30.0-3.fc41. 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [30/63] vim-filesystem-2:9.1.984-1.fc42 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [31/64] perl-Carp-0:1.54-511.fc41.noarc 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [32/65] perl-Exporter-0:5.78-511.fc41.n 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [33/66] perl-overload-0:1.37-512.fc42.n 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [34/67] perl-base-0:2.27-512.fc42.noarc 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [35/68] perl-constant-0:1.33-512.fc41.n 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [36/69] perl-Errno-0:1.38-512.fc42.x86_ 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [37/70] perl-libs-4:5.40.0-512.fc42.x86 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [38/71] perl-Data-Dumper-0:2.189-512.fc 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [39/72] perl-MIME-Base32-0:1.303-21.fc4 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [40/73] perl-MIME-Base64-0:3.16-511.fc4 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [41/74] perl-libnet-0:3.15-512.fc41.noa 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [42/75] perl-parent-1:0.244-1.fc42.noar 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [43/76] less-0:668-1.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [44/77] man-db-0:2.13.0-1.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [45/78] abattis-cantarell-vf-fonts-0:0. 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [46/79] google-noto-sans-vf-fonts-0:202 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [47/80] fonts-filesystem-1:2.0.5-17.fc4 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [48/81] google-noto-fonts-common-0:2024 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [49/84] gnupg2-0:2.4.5-4.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [50/85] libassuan-0:2.5.7-2.fc41.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [51/86] libgpg-error-0:1.51-1.fc42.x86_ 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [52/87] gnutls-0:3.8.8-1.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [53/89] perl-mro-0:1.29-512.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [54/90] perl-overloading-0:0.02-512.fc4 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [55/91] perl-DynaLoader-0:1.56-512.fc42 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [56/92] perl-B-0:1.89-512.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [57/93] perl-Digest-MD5-0:2.59-5.fc41.x 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [58/94] perl-Fcntl-0:1.18-512.fc42.x86_ 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [59/95] perl-FileHandle-0:2.05-512.fc42 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [60/96] perl-IO-0:1.55-512.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [61/97] perl-IO-Socket-IP-0:0.43-1.fc42 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [62/98] perl-POSIX-0:2.20-512.fc42.x86_ 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [63/99] perl-Socket-4:2.038-511.fc41.x8 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 64/100] perl-Symbol-0:1.09-512.fc42.n 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 65/101] perl-Time-Local-2:1.350-511.f 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 66/102] groff-base-0:1.23.0-7.fc41.x8 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 67/103] libpipeline-0:1.5.8-1.fc42.x8 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 68/105] libgcrypt-0:1.11.0-4.fc42.x86 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 69/106] libksba-0:1.6.7-2.fc41.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 70/107] npth-0:1.8-1.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 71/108] tpm2-tss-0:4.1.3-5.fc42.x86_6 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 72/109] nettle-0:3.10-3.fc41.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 73/111] perl-vars-0:1.05-512.fc42.noa 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 74/112] perl-if-0:0.61.000-512.fc42.n 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 75/113] perl-Digest-0:1.20-511.fc41.n 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 76/114] perl-File-stat-0:1.14-512.fc4 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 77/115] perl-SelectSaver-0:1.02-512.f 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 78/116] perl-locale-0:1.12-512.fc42.n 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 79/118] libusb1-0:1.0.27-4.fc42.x86_6 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 80/121] python3-libs-0:3.13.1-2.fc42. 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 81/122] perl-Class-Struct-0:0.68-512. 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 82/123] libb2-0:0.98.1-12.fc41.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 83/124] mpdecimal-0:2.5.1-16.fc41.x86 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 84/125] python-pip-wheel-0:24.3.1-1.f 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 85/126] tzdata-0:2024b-1.fc42.noarch 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 86/127] rocm-runtime-0:6.3.1-1.fc42.x 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 87/128] libdrm-0:2.4.124-1.fc42.x86_6 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 88/129] libpciaccess-0:0.16-13.fc41.x 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 89/130] hwdata-0:0.390-1.fc42.noarch 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 90/131] rocm-hip-0:6.3.1-1.fc42.x86_6 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 91/132] hipcc-0:18-31.rocm6.3.1.fc42. 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 92/133] rocm-comgr-0:18-31.rocm6.3.1. 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 93/134] rocm-device-libs-0:18-31.rocm 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 94/136] clang-resource-filesystem-0:1 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 95/138] llvm-libs-0:19.1.6-2.fc42.x86 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 96/142] gcc-0:14.2.1-6.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 97/143] libstdc++-devel-0:14.2.1-6.fc 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 98/144] cpp-0:14.2.1-6.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [ 99/145] glibc-devel-0:2.40.9000-26.fc 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [100/146] libxcrypt-devel-0:4.4.37-3.fc 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [101/147] perl-Encode-4:3.21-511.fc41.x 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [102/148] perl-Getopt-Long-1:2.58-2.fc4 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [103/149] perl-Storable-1:3.32-511.fc41 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [104/150] perl-Pod-Usage-4:2.03-511.fc4 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [105/151] perl-Text-ParseWords-0:3.31-5 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [106/152] perl-Pod-Perldoc-0:3.28.01-51 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [107/153] perl-podlators-1:6.0.2-2.fc41 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [108/154] perl-File-Temp-1:0.231.100-51 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [109/155] perl-HTTP-Tiny-0:0.090-1.fc42 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [110/156] perl-IPC-Open3-0:1.22-512.fc4 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [111/157] perl-Pod-Simple-1:3.45-511.fc 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [112/158] perl-Term-ANSIColor-0:5.01-51 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [113/159] perl-Term-Cap-0:1.18-511.fc41 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [114/160] perl-File-Path-0:2.18-511.fc4 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [115/161] perl-IO-Socket-SSL-0:2.089-1. 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [116/162] perl-Net-SSLeay-0:1.94-7.fc41 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [117/163] perl-Pod-Escapes-1:1.07-511.f 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [118/164] perl-Text-Tabs+Wrap-0:2024.00 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [119/165] ncurses-0:6.5-2.20240629.fc41 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [120/166] perl-AutoLoader-0:5.74-512.fc 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [121/167] libgfortran-0:14.2.1-6.fc42.x 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [122/168] libquadmath-0:14.2.1-6.fc42.x 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [123/169] systemd-0:257.1-1.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [124/170] dbus-1:1.16.0-1.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [125/171] libseccomp-0:2.5.5-2.fc41.x86 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [126/172] systemd-pam-0:257.1-1.fc42.x8 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [127/173] dbus-broker-0:36-4.fc41.x86_6 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [128/174] dbus-common-1:1.16.0-1.fc42.n 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [129/191] kernel-headers-0:6.13.0-0.rc5 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [130/192] procps-ng-0:4.0.4-4.fc41.x86_ 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [131/193] tcl-1:8.6.15-6.fc42.x86_64 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [132/194] rocm-clang-devel-0:18-31.rocm 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [133/195] rocm-lld-0:18-31.rocm6.3.1.fc 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [134/196] rocm-llvm-static-0:18-31.rocm 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [135/197] rocm-clang-0:18-31.rocm6.3.1. 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [136/198] rocm-clang-libs-0:18-31.rocm6 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [137/199] python3-0:3.13.1-2.fc42.x86_6 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [138/200] rocm-clang-runtime-devel-0:18 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [139/201] rocm-llvm-libs-0:18-31.rocm6. 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [140/203] pcre2-utf16-0:10.44-1.fc41.1. 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [141/205] rocm-llvm-devel-0:18-31.rocm6 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [142/206] rocm-llvm-0:18-31.rocm6.3.1.f 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [143/207] annobin-plugin-gcc-0:12.80-1. 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [144/208] annobin-docs-0:12.80-1.fc42.n 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [145/209] systemd-rpm-macros-0:257.1-1. 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [146/210] cmake-rpm-macros-0:3.31.3-1.f 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [147/211] gcc-plugin-annobin-0:14.2.1-6 100% | 0.0 B/s | 0.0 B | 00m00s >>> Already downloaded [148/211] langpacks-en-0:4.2-2.fc42.noa 100% | 44.1 KiB/s | 10.9 KiB | 00m00s [149/211] wget2-wget-0:2.2.0-1.fc42.x86 100% | 37.6 KiB/s | 9.6 KiB | 00m00s [150/211] hipblas-devel-0:6.3.0-1.fc42. 100% | 1.4 MiB/s | 93.9 KiB | 00m00s [151/211] hipcc-libomp-devel-0:18-31.ro 100% | 404.7 KiB/s | 13.0 KiB | 00m00s [152/211] xxd-2:9.1.984-1.fc42.x86_64 100% | 235.9 KiB/s | 34.7 KiB | 00m00s [153/211] pthreadpool-devel-0:0.0^git20 100% | 174.0 KiB/s | 14.3 KiB | 00m00s [154/211] rocblas-devel-0:6.3.0-2.fc42. 100% | 5.1 MiB/s | 98.4 KiB | 00m00s [155/211] rocm-rpm-macros-modules-0:6.3 100% | 224.6 KiB/s | 18.9 KiB | 00m00s [156/211] langpacks-core-en-0:4.2-2.fc4 100% | 136.3 KiB/s | 10.9 KiB | 00m00s [157/211] langpacks-fonts-en-0:4.2-2.fc 100% | 140.2 KiB/s | 11.2 KiB | 00m00s [158/211] openmpi-0:5.0.6-1.fc42.x86_64 100% | 2.7 MiB/s | 2.0 MiB | 00m01s [159/211] libcurl-devel-0:8.11.1-2.fc42 100% | 1.9 MiB/s | 883.7 KiB | 00m00s [160/211] libpsm2-0:12.0.1-1.fc42.x86_6 100% | 2.5 MiB/s | 200.0 KiB | 00m00s [161/211] openssh-clients-0:9.9p1-5.fc4 100% | 6.3 MiB/s | 761.6 KiB | 00m00s [162/211] orangefs-0:2.9.8-12.fc41.x86_ 100% | 15.3 MiB/s | 1.8 MiB | 00m00s [163/211] prrte-0:3.0.6-1.fc42.x86_64 100% | 767.4 KiB/s | 56.8 KiB | 00m00s [164/211] pmix-0:4.2.8-3.fc41.x86_64 100% | 5.9 MiB/s | 668.9 KiB | 00m00s [165/211] ucx-0:1.17.0-3.fc42.x86_64 100% | 8.9 MiB/s | 835.4 KiB | 00m00s [166/211] wget2-0:2.2.0-1.fc42.x86_64 100% | 2.9 MiB/s | 277.5 KiB | 00m00s [167/211] libfabric-0:1.22.0-1.fc41.x86 100% | 3.5 MiB/s | 1.4 MiB | 00m00s [168/211] google-noto-sans-mono-vf-font 100% | 1.6 MiB/s | 278.2 KiB | 00m00s [169/211] pthreadpool-0:0.0^git20230829 100% | 198.6 KiB/s | 46.1 KiB | 00m00s [170/211] libibverbs-0:54.0-3.fc42.x86_ 100% | 4.9 MiB/s | 440.1 KiB | 00m00s [171/211] libnl3-0:3.11.0-1.fc42.x86_64 100% | 3.9 MiB/s | 353.1 KiB | 00m00s [172/211] google-noto-serif-vf-fonts-0: 100% | 2.5 MiB/s | 645.8 KiB | 00m00s [173/211] librdmacm-0:54.0-3.fc42.x86_6 100% | 842.1 KiB/s | 70.7 KiB | 00m00s [174/211] openssh-0:9.9p1-5.fc42.x86_64 100% | 3.9 MiB/s | 353.3 KiB | 00m00s [175/211] libfido2-0:1.15.0-2.fc41.x86_ 100% | 598.4 KiB/s | 98.1 KiB | 00m00s [176/211] munge-libs-0:0.5.16-3.fc41.x8 100% | 260.9 KiB/s | 21.4 KiB | 00m00s [177/211] prrte-libs-0:3.0.6-1.fc42.x86 100% | 4.9 MiB/s | 550.0 KiB | 00m00s [178/211] tcsh-0:6.24.14-1.fc42.x86_64 100% | 1.8 MiB/s | 455.3 KiB | 00m00s [179/211] wget2-libs-0:2.2.0-1.fc42.x86 100% | 1.7 MiB/s | 143.3 KiB | 00m00s [180/211] libcbor-0:0.11.0-2.fc41.x86_6 100% | 409.1 KiB/s | 33.1 KiB | 00m00s [181/211] torque-libs-0:6.1.3-13.fc42.x 100% | 2.1 MiB/s | 187.9 KiB | 00m00s [182/211] gnutls-dane-0:3.8.8-1.fc42.x8 100% | 523.9 KiB/s | 43.5 KiB | 00m00s [183/211] munge-0:0.5.16-3.fc41.x86_64 100% | 1.6 MiB/s | 133.8 KiB | 00m00s [184/211] logrotate-0:3.22.0-2.fc41.x86 100% | 1.0 MiB/s | 75.9 KiB | 00m00s [185/211] hiredis-0:1.2.0-4.fc42.x86_64 100% | 657.2 KiB/s | 49.9 KiB | 00m00s [186/211] unbound-libs-0:1.22.0-8.fc42. 100% | 2.2 MiB/s | 554.2 KiB | 00m00s [187/211] protobuf-c-0:1.5.0-4.fc41.x86 100% | 420.7 KiB/s | 32.4 KiB | 00m00s [188/211] libomp-0:19.1.6-2.fc42.x86_64 100% | 3.3 MiB/s | 628.3 KiB | 00m00s [189/211] libomp-devel-0:19.1.6-2.fc42. 100% | 2.8 MiB/s | 655.5 KiB | 00m00s [190/211] hipblas-common-devel-0:6.3.0- 100% | 165.8 KiB/s | 13.3 KiB | 00m00s [191/211] hipblas-0:6.3.0-1.fc42.x86_64 100% | 1.3 MiB/s | 170.6 KiB | 00m00s [192/211] brotli-devel-0:1.1.0-5.fc41.x 100% | 373.7 KiB/s | 34.0 KiB | 00m00s [193/211] brotli-0:1.1.0-5.fc41.x86_64 100% | 233.5 KiB/s | 20.1 KiB | 00m00s [194/211] krb5-devel-0:1.21.3-3.fc42.x8 100% | 858.3 KiB/s | 143.3 KiB | 00m00s [195/211] libkadm5-0:1.21.3-3.fc42.x86_ 100% | 465.8 KiB/s | 76.9 KiB | 00m00s [196/211] libidn2-devel-0:2.3.7-2.fc41. 100% | 700.8 KiB/s | 70.8 KiB | 00m00s [197/211] rocblas-0:6.3.0-2.fc42.x86_64 100% | 81.6 MiB/s | 203.2 MiB | 00m02s [198/211] libnghttp2-devel-0:1.64.0-1.f 100% | 194.2 KiB/s | 55.7 KiB | 00m00s [199/211] libpsl-devel-0:0.21.5-4.fc41. 100% | 386.8 KiB/s | 33.3 KiB | 00m00s [200/211] publicsuffix-list-0:20240107- 100% | 961.5 KiB/s | 87.5 KiB | 00m00s [201/211] libssh-devel-0:0.11.1-1.fc42. 100% | 469.0 KiB/s | 42.2 KiB | 00m00s [202/211] zlib-ng-compat-devel-0:2.2.2- 100% | 414.8 KiB/s | 38.2 KiB | 00m00s [203/211] keyutils-libs-devel-0:1.6.3-4 100% | 332.8 KiB/s | 59.9 KiB | 00m00s [204/211] libcom_err-devel-0:1.47.1-6.f 100% | 181.6 KiB/s | 16.3 KiB | 00m00s [205/211] rocsolver-0:6.3.0-1.fc42.x86_ 100% | 58.4 MiB/s | 98.0 MiB | 00m02s [206/211] libselinux-devel-0:3.8-0.rc3. 100% | 815.7 KiB/s | 151.7 KiB | 00m00s [207/211] openssl-devel-1:3.2.2-10.fc42 100% | 4.4 MiB/s | 2.8 MiB | 00m01s [208/211] libsepol-devel-0:3.8-0.rc3.1. 100% | 479.8 KiB/s | 48.5 KiB | 00m00s [209/211] pcre2-devel-0:10.44-1.fc41.1. 100% | 3.3 MiB/s | 521.4 KiB | 00m00s [210/211] libverto-devel-0:0.3.2-9.fc41 100% | 59.0 KiB/s | 14.4 KiB | 00m00s [211/211] pcre2-utf32-0:10.44-1.fc41.1. 100% | 1.1 MiB/s | 210.4 KiB | 00m00s -------------------------------------------------------------------------------- [211/211] Total 100% | 63.6 MiB/s | 320.6 MiB | 00m05s Running transaction [ 1/213] Verify package files 100% | 86.0 B/s | 211.0 B | 00m02s [ 2/213] Prepare transaction 100% | 802.0 B/s | 211.0 B | 00m00s [ 3/213] Installing cmake-filesystem-0 100% | 2.5 MiB/s | 7.6 KiB | 00m00s [ 4/213] Installing rocm-llvm-libs-0:1 100% | 65.4 MiB/s | 105.0 MiB | 00m02s [ 5/213] Installing libgpg-error-0:1.5 100% | 145.4 MiB/s | 893.1 KiB | 00m00s [ 6/213] Installing hwloc-libs-0:2.11. 100% | 358.3 MiB/s | 2.9 MiB | 00m00s [ 7/213] Installing fonts-filesystem-1 100% | 0.0 B/s | 788.0 B | 00m00s [ 8/213] Installing numactl-libs-0:2.0 100% | 54.0 MiB/s | 55.3 KiB | 00m00s [ 9/213] Installing google-noto-fonts- 100% | 0.0 B/s | 18.3 KiB | 00m00s [ 10/213] Installing munge-libs-0:0.5.1 100% | 32.2 MiB/s | 32.9 KiB | 00m00s [ 11/213] Installing pmix-0:4.2.8-3.fc4 100% | 224.0 MiB/s | 2.0 MiB | 00m00s [ 12/213] Installing libnl3-0:3.11.0-1. 100% | 171.1 MiB/s | 1.0 MiB | 00m00s [ 13/213] Installing libibverbs-0:54.0- 100% | 136.5 MiB/s | 1.2 MiB | 00m00s [ 14/213] Installing libmpc-0:1.3.1-6.f 100% | 162.3 MiB/s | 166.2 KiB | 00m00s [ 15/213] Installing expat-0:2.6.4-1.fc 100% | 140.4 MiB/s | 287.6 KiB | 00m00s [ 16/213] Installing libpsm2-0:12.0.1-1 100% | 143.6 MiB/s | 441.1 KiB | 00m00s [ 17/213] Installing libassuan-0:2.5.7- 100% | 80.9 MiB/s | 165.6 KiB | 00m00s [ 18/213] Installing rocm-clang-libs-0: 100% | 68.6 MiB/s | 116.3 MiB | 00m02s [ 19/213] Installing rocm-comgr-0:18-31 100% | 65.0 MiB/s | 150.2 MiB | 00m02s [ 20/213] Installing nettle-0:3.10-3.fc 100% | 194.4 MiB/s | 796.1 KiB | 00m00s [ 21/213] Installing gnutls-0:3.8.8-1.f 100% | 125.0 MiB/s | 3.2 MiB | 00m00s [ 22/213] Installing groff-base-0:1.23. 100% | 107.6 MiB/s | 3.9 MiB | 00m00s [ 23/213] Installing less-0:668-1.fc42. 100% | 133.4 MiB/s | 409.7 KiB | 00m00s [ 24/213] Installing vim-filesystem-2:9 100% | 2.3 MiB/s | 4.7 KiB | 00m00s [ 25/213] Installing libedit-0:3.1-54.2 100% | 117.7 MiB/s | 241.0 KiB | 00m00s [ 26/213] Installing make-1:4.4.1-9.fc4 100% | 180.0 MiB/s | 1.8 MiB | 00m00s [ 27/213] Installing jsoncpp-0:1.9.5-8. 100% | 124.5 MiB/s | 254.9 KiB | 00m00s [ 28/213] Installing llvm-libs-0:19.1.6 100% | 341.6 MiB/s | 123.0 MiB | 00m00s [ 29/213] Installing libomp-0:19.1.6-2. 100% | 258.9 MiB/s | 1.8 MiB | 00m00s [ 30/213] Installing cpp-0:14.2.1-6.fc4 100% | 305.8 MiB/s | 34.6 MiB | 00m00s [ 31/213] Installing librdmacm-0:54.0-3 100% | 76.7 MiB/s | 157.0 KiB | 00m00s [ 32/213] Installing libfabric-0:1.22.0 100% | 344.3 MiB/s | 5.2 MiB | 00m00s [ 33/213] Installing google-noto-sans-m 100% | 183.0 MiB/s | 562.2 KiB | 00m00s [ 34/213] Installing google-noto-serif- 100% | 253.3 MiB/s | 1.5 MiB | 00m00s [ 35/213] Installing google-noto-sans-v 100% | 249.8 MiB/s | 1.2 MiB | 00m00s [ 36/213] Installing abattis-cantarell- 100% | 94.9 MiB/s | 194.4 KiB | 00m00s [ 37/213] Installing default-fonts-core 100% | 8.9 MiB/s | 18.2 KiB | 00m00s [ 38/213] Installing langpacks-core-en- 100% | 0.0 B/s | 704.0 B | 00m00s [ 39/213] Installing langpacks-fonts-en 100% | 0.0 B/s | 652.0 B | 00m00s [ 40/213] Installing libgcrypt-0:1.11.0 100% | 258.3 MiB/s | 1.6 MiB | 00m00s [ 41/213] Installing libksba-0:1.6.7-2. 100% | 130.5 MiB/s | 401.0 KiB | 00m00s [ 42/213] Installing rocm-lld-0:18-31.r 100% | 65.8 MiB/s | 7.4 MiB | 00m00s [ 43/213] Installing hipblas-common-dev 100% | 17.4 MiB/s | 17.8 KiB | 00m00s [ 44/213] Installing libssh-devel-0:0.1 100% | 88.0 MiB/s | 180.3 KiB | 00m00s [ 45/213] Installing zlib-ng-compat-dev 100% | 105.8 MiB/s | 108.3 KiB | 00m00s [ 46/213] Installing annobin-docs-0:12. 100% | 97.4 MiB/s | 99.7 KiB | 00m00s [ 47/213] Installing pcre2-utf32-0:10.4 100% | 274.8 MiB/s | 562.8 KiB | 00m00s [ 48/213] Installing pcre2-utf16-0:10.4 100% | 192.4 MiB/s | 590.9 KiB | 00m00s [ 49/213] Installing pcre2-devel-0:10.4 100% | 153.5 MiB/s | 2.0 MiB | 00m00s [ 50/213] Installing rocm-clang-runtime 100% | 106.0 MiB/s | 6.9 MiB | 00m00s [ 51/213] Installing tcl-1:8.6.15-6.fc4 100% | 193.2 MiB/s | 4.2 MiB | 00m00s [ 52/213] Installing procps-ng-0:4.0.4- 100% | 104.1 MiB/s | 1.0 MiB | 00m00s [ 53/213] Installing kernel-headers-0:6 100% | 116.6 MiB/s | 6.6 MiB | 00m00s [ 54/213] Installing libxcrypt-devel-0: 100% | 10.7 MiB/s | 32.9 KiB | 00m00s [ 55/213] Installing glibc-devel-0:2.40 100% | 86.4 MiB/s | 2.3 MiB | 00m00s [ 56/213] Installing gcc-0:14.2.1-6.fc4 100% | 317.3 MiB/s | 103.4 MiB | 00m00s [ 57/213] Installing libverto-devel-0:0 100% | 25.7 MiB/s | 26.4 KiB | 00m00s [ 58/213] Installing libsepol-devel-0:3 100% | 31.3 MiB/s | 128.3 KiB | 00m00s [ 59/213] Installing libselinux-devel-0 100% | 17.5 MiB/s | 161.6 KiB | 00m00s [ 60/213] Installing libcom_err-devel-0 100% | 17.8 MiB/s | 18.3 KiB | 00m00s [ 61/213] Installing keyutils-libs-deve 100% | 6.0 MiB/s | 55.2 KiB | 00m00s [ 62/213] Installing openssl-devel-1:3. 100% | 30.1 MiB/s | 5.2 MiB | 00m00s [ 63/213] Installing publicsuffix-list- 100% | 311.6 MiB/s | 319.1 KiB | 00m00s [ 64/213] Installing libpsl-devel-0:0.2 100% | 55.5 MiB/s | 113.6 KiB | 00m00s [ 65/213] Installing libnghttp2-devel-0 100% | 289.5 MiB/s | 296.5 KiB | 00m00s [ 66/213] Installing libidn2-devel-0:2. 100% | 84.7 MiB/s | 260.1 KiB | 00m00s [ 67/213] Installing libkadm5-0:1.21.3- 100% | 104.3 MiB/s | 213.6 KiB | 00m00s [ 68/213] Installing krb5-devel-0:1.21. 100% | 139.7 MiB/s | 715.2 KiB | 00m00s [ 69/213] Installing brotli-0:1.1.0-5.f 100% | 31.7 MiB/s | 32.5 KiB | 00m00s [ 70/213] Installing brotli-devel-0:1.1 100% | 66.4 MiB/s | 68.0 KiB | 00m00s [ 71/213] Installing dbus-common-1:1.16 100% | 338.8 KiB/s | 13.6 KiB | 00m00s [ 72/213] Installing dbus-broker-0:36-4 100% | 26.9 MiB/s | 385.3 KiB | 00m00s [ 73/213] Installing dbus-1:1.16.0-1.fc 100% | 0.0 B/s | 124.0 B | 00m00s [ 74/213] Installing libseccomp-0:2.5.5 100% | 85.5 MiB/s | 175.2 KiB | 00m00s [ 75/213] Installing systemd-pam-0:257. 100% | 137.4 MiB/s | 1.1 MiB | 00m00s [ 76/213] Installing systemd-0:257.1-1. 100% | 60.5 MiB/s | 17.5 MiB | 00m00s >>> Running post-install scriptlet: systemd-0:257.1-1.fc42.x86_64 >>> Finished post-install scriptlet: systemd-0:257.1-1.fc42.x86_64 >>> Scriptlet output: >>> Creating group 'systemd-journal' with GID 190. >>> Creating group 'systemd-oom' with GID 999. >>> Creating user 'systemd-oom' (systemd Userspace OOM Killer) with UID 999 and >>> [ 77/213] Installing logrotate-0:3.22.0 100% | 3.2 MiB/s | 155.7 KiB | 00m00s >>> Running post-install scriptlet: logrotate-0:3.22.0-2.fc41.x86_64 >>> Finished post-install scriptlet: logrotate-0:3.22.0-2.fc41.x86_64 >>> Scriptlet output: >>> Created symlink '/etc/systemd/system/timers.target.wants/logrotate.timer' ↠>>> [ 78/213] Installing munge-0:0.5.16-3.f 100% | 13.8 MiB/s | 352.4 KiB | 00m00s [ 79/213] Installing torque-libs-0:6.1. 100% | 149.5 MiB/s | 459.1 KiB | 00m00s [ 80/213] Installing prrte-libs-0:3.0.6 100% | 153.4 MiB/s | 1.7 MiB | 00m00s [ 81/213] Installing prrte-0:3.0.6-1.fc 100% | 58.2 MiB/s | 178.8 KiB | 00m00s [ 82/213] Installing libquadmath-0:14.2 100% | 158.6 MiB/s | 324.7 KiB | 00m00s [ 83/213] Installing libgfortran-0:14.2 100% | 302.7 MiB/s | 3.0 MiB | 00m00s [ 84/213] Installing ncurses-0:6.5-2.20 100% | 154.8 MiB/s | 633.9 KiB | 00m00s [ 85/213] Installing perl-Digest-0:1.20 100% | 36.2 MiB/s | 37.1 KiB | 00m00s [ 86/213] Installing perl-B-0:1.89-512. 100% | 163.2 MiB/s | 501.3 KiB | 00m00s [ 87/213] Installing perl-Digest-MD5-0: 100% | 60.2 MiB/s | 61.7 KiB | 00m00s [ 88/213] Installing perl-FileHandle-0: 100% | 0.0 B/s | 9.8 KiB | 00m00s [ 89/213] Installing perl-MIME-Base32-0 100% | 31.4 MiB/s | 32.2 KiB | 00m00s [ 90/213] Installing perl-Data-Dumper-0 100% | 55.5 MiB/s | 113.6 KiB | 00m00s [ 91/213] Installing perl-libnet-0:3.15 100% | 95.9 MiB/s | 294.7 KiB | 00m00s [ 92/213] Installing perl-URI-0:5.31-1. 100% | 52.7 MiB/s | 269.6 KiB | 00m00s [ 93/213] Installing perl-IO-Socket-IP- 100% | 99.8 MiB/s | 102.2 KiB | 00m00s [ 94/213] Installing perl-AutoLoader-0: 100% | 20.5 MiB/s | 20.9 KiB | 00m00s [ 95/213] Installing perl-Time-Local-2: 100% | 68.9 MiB/s | 70.6 KiB | 00m00s [ 96/213] Installing perl-if-0:0.61.000 100% | 0.0 B/s | 6.2 KiB | 00m00s [ 97/213] Installing perl-locale-0:1.12 100% | 0.0 B/s | 6.9 KiB | 00m00s [ 98/213] Installing perl-Pod-Escapes-1 100% | 25.3 MiB/s | 25.9 KiB | 00m00s [ 99/213] Installing perl-Text-Tabs+Wra 100% | 23.3 MiB/s | 23.9 KiB | 00m00s [100/213] Installing perl-IO-Socket-SSL 100% | 172.7 MiB/s | 707.4 KiB | 00m00s [101/213] Installing perl-Net-SSLeay-0: 100% | 151.4 MiB/s | 1.4 MiB | 00m00s [102/213] Installing perl-File-Path-0:2 100% | 31.5 MiB/s | 64.5 KiB | 00m00s [103/213] Installing perl-POSIX-0:2.20- 100% | 115.4 MiB/s | 236.4 KiB | 00m00s [104/213] Installing perl-Class-Struct- 100% | 25.3 MiB/s | 25.9 KiB | 00m00s [105/213] Installing perl-IPC-Open3-0:1 100% | 22.7 MiB/s | 23.3 KiB | 00m00s [106/213] Installing perl-Term-ANSIColo 100% | 96.9 MiB/s | 99.2 KiB | 00m00s [107/213] Installing perl-Term-Cap-0:1. 100% | 29.9 MiB/s | 30.6 KiB | 00m00s [108/213] Installing perl-HTTP-Tiny-0:0 100% | 76.4 MiB/s | 156.4 KiB | 00m00s [109/213] Installing perl-Pod-Simple-1: 100% | 139.3 MiB/s | 570.5 KiB | 00m00s [110/213] Installing perl-File-Temp-1:0 100% | 80.1 MiB/s | 164.1 KiB | 00m00s [111/213] Installing perl-Socket-4:2.03 100% | 123.1 MiB/s | 126.1 KiB | 00m00s [112/213] Installing perl-Symbol-0:1.09 100% | 0.0 B/s | 7.2 KiB | 00m00s [113/213] Installing perl-SelectSaver-0 100% | 0.0 B/s | 2.6 KiB | 00m00s [114/213] Installing perl-File-stat-0:1 100% | 12.7 MiB/s | 13.1 KiB | 00m00s [115/213] Installing perl-podlators-1:6 100% | 104.6 MiB/s | 321.4 KiB | 00m00s [116/213] Installing perl-Pod-Perldoc-0 100% | 55.1 MiB/s | 169.3 KiB | 00m00s [117/213] Installing perl-base-0:2.27-5 100% | 0.0 B/s | 12.9 KiB | 00m00s [118/213] Installing perl-mro-0:1.29-51 100% | 45.6 MiB/s | 46.7 KiB | 00m00s [119/213] Installing perl-overloading-0 100% | 5.4 MiB/s | 5.5 KiB | 00m00s [120/213] Installing perl-Fcntl-0:1.18- 100% | 48.9 MiB/s | 50.1 KiB | 00m00s [121/213] Installing perl-IO-0:1.55-512 100% | 75.8 MiB/s | 155.2 KiB | 00m00s [122/213] Installing perl-Text-ParseWor 100% | 14.2 MiB/s | 14.6 KiB | 00m00s [123/213] Installing perl-Pod-Usage-4:2 100% | 84.3 MiB/s | 86.3 KiB | 00m00s [124/213] Installing perl-Getopt-Std-0: 100% | 11.5 MiB/s | 11.7 KiB | 00m00s [125/213] Installing perl-Scalar-List-U 100% | 74.5 MiB/s | 152.6 KiB | 00m00s [126/213] Installing perl-constant-0:1. 100% | 26.7 MiB/s | 27.4 KiB | 00m00s [127/213] Installing perl-Errno-0:1.38- 100% | 0.0 B/s | 8.8 KiB | 00m00s [128/213] Installing perl-MIME-Base64-0 100% | 23.6 MiB/s | 48.4 KiB | 00m00s [129/213] Installing perl-parent-1:0.24 100% | 10.7 MiB/s | 11.0 KiB | 00m00s [130/213] Installing perl-overload-0:1. 100% | 70.3 MiB/s | 71.9 KiB | 00m00s [131/213] Installing perl-vars-0:1.05-5 100% | 0.0 B/s | 4.3 KiB | 00m00s [132/213] Installing perl-Storable-1:3. 100% | 114.3 MiB/s | 234.0 KiB | 00m00s [133/213] Installing perl-Getopt-Long-1 100% | 71.9 MiB/s | 147.2 KiB | 00m00s [134/213] Installing perl-File-Basename 100% | 0.0 B/s | 14.6 KiB | 00m00s [135/213] Installing perl-Carp-0:1.54-5 100% | 46.6 MiB/s | 47.7 KiB | 00m00s [136/213] Installing perl-Exporter-0:5. 100% | 54.3 MiB/s | 55.6 KiB | 00m00s [137/213] Installing perl-PathTools-0:3 100% | 60.1 MiB/s | 184.6 KiB | 00m00s [138/213] Installing perl-DynaLoader-0: 100% | 31.7 MiB/s | 32.5 KiB | 00m00s [139/213] Installing perl-Encode-4:3.21 100% | 235.9 MiB/s | 4.7 MiB | 00m00s [140/213] Installing perl-libs-4:5.40.0 100% | 153.5 MiB/s | 10.0 MiB | 00m00s [141/213] Installing perl-interpreter-4 100% | 121.1 MiB/s | 124.0 KiB | 00m00s [142/213] Installing perl-File-Copy-0:2 100% | 19.7 MiB/s | 20.2 KiB | 00m00s [143/213] Installing perl-File-Which-0: 100% | 15.3 MiB/s | 31.4 KiB | 00m00s [144/213] Installing libstdc++-devel-0: 100% | 219.1 MiB/s | 15.6 MiB | 00m00s [145/213] Installing gcc-c++-0:14.2.1-6 100% | 311.3 MiB/s | 37.7 MiB | 00m00s [146/213] Installing clang-resource-fil 100% | 16.3 MiB/s | 16.7 KiB | 00m00s [147/213] Installing libomp-devel-0:19. 100% | 623.3 MiB/s | 31.2 MiB | 00m00s [148/213] Installing hwdata-0:0.390-1.f 100% | 406.1 MiB/s | 9.3 MiB | 00m00s [149/213] Installing libpciaccess-0:0.1 100% | 44.9 MiB/s | 46.0 KiB | 00m00s [150/213] Installing libdrm-0:2.4.124-1 100% | 130.7 MiB/s | 401.6 KiB | 00m00s [151/213] Installing rocm-runtime-0:6.3 100% | 314.1 MiB/s | 2.8 MiB | 00m00s [152/213] Installing rocm-runtime-devel 100% | 92.6 MiB/s | 569.2 KiB | 00m00s [153/213] Installing tzdata-0:2024b-1.f 100% | 25.5 MiB/s | 1.9 MiB | 00m00s [154/213] Installing python-pip-wheel-0 100% | 414.7 MiB/s | 1.2 MiB | 00m00s [155/213] Installing mpdecimal-0:2.5.1- 100% | 100.6 MiB/s | 206.0 KiB | 00m00s [156/213] Installing libb2-0:0.98.1-12. 100% | 6.0 MiB/s | 43.3 KiB | 00m00s [157/213] Installing python3-libs-0:3.1 100% | 198.1 MiB/s | 40.2 MiB | 00m00s [158/213] Installing python3-0:3.13.1-2 100% | 31.6 MiB/s | 32.4 KiB | 00m00s [159/213] Installing cmake-rpm-macros-0 100% | 8.0 MiB/s | 8.2 KiB | 00m00s [160/213] Installing rocm-clang-0:18-31 100% | 71.3 MiB/s | 125.2 MiB | 00m02s [161/213] Installing rocm-clang-devel-0 100% | 94.3 MiB/s | 22.2 MiB | 00m00s [162/213] Installing rocm-llvm-0:18-31. 100% | 67.8 MiB/s | 87.5 MiB | 00m01s [163/213] Installing rocm-llvm-devel-0: 100% | 71.0 MiB/s | 24.7 MiB | 00m00s [164/213] Installing rocm-llvm-static-0 100% | 88.8 MiB/s | 241.4 MiB | 00m03s [165/213] Installing rocm-device-libs-0 100% | 78.9 MiB/s | 3.2 MiB | 00m00s [166/213] Installing hipcc-0:18-31.rocm 100% | 75.4 MiB/s | 772.1 KiB | 00m00s [167/213] Installing rocm-hip-0:6.3.1-1 100% | 313.4 MiB/s | 23.2 MiB | 00m00s [168/213] Installing rocm-comgr-devel-0 100% | 50.9 MiB/s | 104.3 KiB | 00m00s [169/213] Installing protobuf-c-0:1.5.0 100% | 54.2 MiB/s | 55.5 KiB | 00m00s [170/213] Installing hiredis-0:1.2.0-4. 100% | 4.9 MiB/s | 115.7 KiB | 00m00s [171/213] Installing unbound-libs-0:1.2 100% | 179.7 MiB/s | 1.4 MiB | 00m00s [172/213] Installing gnutls-dane-0:3.8. 100% | 69.2 MiB/s | 70.8 KiB | 00m00s [173/213] Installing libusb1-0:1.0.27-4 100% | 7.0 MiB/s | 172.3 KiB | 00m00s [174/213] Installing tpm2-tss-0:4.1.3-5 100% | 158.2 MiB/s | 1.6 MiB | 00m00s [175/213] Installing npth-0:1.8-1.fc42. 100% | 53.4 MiB/s | 54.7 KiB | 00m00s [176/213] Installing gnupg2-0:2.4.5-4.f 100% | 239.7 MiB/s | 9.6 MiB | 00m00s [177/213] Installing gpgme-0:1.24.0-1.f 100% | 191.8 MiB/s | 589.3 KiB | 00m00s [178/213] Installing wget2-libs-0:2.2.0 100% | 178.7 MiB/s | 366.0 KiB | 00m00s [179/213] Installing wget2-0:2.2.0-1.fc 100% | 150.3 MiB/s | 1.1 MiB | 00m00s [180/213] Installing libpipeline-0:1.5. 100% | 8.2 MiB/s | 150.6 KiB | 00m00s [181/213] Installing man-db-0:2.13.0-1. 100% | 72.3 MiB/s | 2.9 MiB | 00m00s [182/213] Installing environment-module 100% | 69.4 MiB/s | 1.8 MiB | 00m00s [183/213] Installing rocm-rpm-macros-mo 100% | 3.8 MiB/s | 31.5 KiB | 00m00s [184/213] Installing rocblas-0:6.3.0-2. 100% | 599.8 MiB/s | 3.7 GiB | 00m06s [185/213] Installing rocsolver-0:6.3.0- 100% | 542.7 MiB/s | 117.8 MiB | 00m00s [186/213] Installing hipblas-0:6.3.0-1. 100% | 375.1 MiB/s | 1.1 MiB | 00m00s [187/213] Installing libcbor-0:0.11.0-2 100% | 73.5 MiB/s | 75.3 KiB | 00m00s [188/213] Installing libfido2-0:1.15.0- 100% | 117.1 MiB/s | 239.7 KiB | 00m00s [189/213] Installing emacs-filesystem-1 100% | 0.0 B/s | 544.0 B | 00m00s [190/213] Installing tcsh-0:6.24.14-1.f 100% | 54.1 MiB/s | 1.2 MiB | 00m00s [191/213] Installing orangefs-0:2.9.8-1 100% | 183.0 MiB/s | 3.1 MiB | 00m00s [192/213] Installing openssh-0:9.9p1-5. 100% | 229.9 MiB/s | 1.4 MiB | 00m00s [193/213] Installing openssh-clients-0: 100% | 76.0 MiB/s | 2.7 MiB | 00m00s [194/213] Installing pthreadpool-0:0.0^ 100% | 111.6 MiB/s | 114.3 KiB | 00m00s [195/213] Installing rhash-0:1.4.5-1.fc 100% | 118.7 MiB/s | 364.6 KiB | 00m00s [196/213] Installing libuv-1:1.49.2-1.f 100% | 46.6 MiB/s | 572.2 KiB | 00m00s [197/213] Installing cmake-data-0:3.31. 100% | 55.2 MiB/s | 9.1 MiB | 00m00s [198/213] Installing cmake-0:3.31.3-1.f 100% | 313.8 MiB/s | 32.9 MiB | 00m00s [199/213] Installing ucx-0:1.17.0-3.fc4 100% | 215.1 MiB/s | 2.4 MiB | 00m00s [200/213] Installing openmpi-0:5.0.6-1. 100% | 270.0 MiB/s | 7.0 MiB | 00m00s [201/213] Installing pthreadpool-devel- 100% | 97.5 MiB/s | 99.8 KiB | 00m00s [202/213] Installing hipblas-devel-0:6. 100% | 519.0 MiB/s | 3.1 MiB | 00m00s [203/213] Installing rocblas-devel-0:6. 100% | 465.4 MiB/s | 2.8 MiB | 00m00s [204/213] Installing rocm-rpm-macros-0: 100% | 0.0 B/s | 19.3 KiB | 00m00s [205/213] Installing wget2-wget-0:2.2.0 100% | 0.0 B/s | 444.0 B | 00m00s [206/213] Installing rocm-hip-devel-0:6 100% | 268.8 MiB/s | 2.7 MiB | 00m00s [207/213] Installing hipcc-libomp-devel 100% | 60.5 KiB/s | 124.0 B | 00m00s [208/213] Installing libcurl-devel-0:8. 100% | 50.4 MiB/s | 1.4 MiB | 00m00s [209/213] Installing annobin-plugin-gcc 100% | 40.5 MiB/s | 994.5 KiB | 00m00s [210/213] Installing gcc-plugin-annobin 100% | 2.5 MiB/s | 61.2 KiB | 00m00s [211/213] Installing langpacks-en-0:4.2 100% | 683.6 KiB/s | 700.0 B | 00m00s [212/213] Installing systemd-rpm-macros 100% | 11.0 MiB/s | 11.3 KiB | 00m00s [213/213] Installing xxd-2:9.1.984-1.fc 100% | 120.4 KiB/s | 44.9 KiB | 00m00s Warning: skipped OpenPGP checks for 31 packages from repository: copr_base Complete! Finish: build setup for llama-cpp-b4094-8.fc42.src.rpm Start: rpmbuild llama-cpp-b4094-8.fc42.src.rpm Building target platforms: x86_64 Building for target x86_64 setting SOURCE_DATE_EPOCH=1733184000 Executing(%mkbuilddir): /bin/sh -e /var/tmp/rpm-tmp.lNtcOy + umask 022 + cd /builddir/build/BUILD/llama-cpp-b4094-build + test -d /builddir/build/BUILD/llama-cpp-b4094-build + /usr/bin/chmod -Rf a+rX,u+w,g-w,o-w /builddir/build/BUILD/llama-cpp-b4094-build + /usr/bin/rm -rf /builddir/build/BUILD/llama-cpp-b4094-build + /usr/bin/mkdir -p /builddir/build/BUILD/llama-cpp-b4094-build + /usr/bin/mkdir -p /builddir/build/BUILD/llama-cpp-b4094-build/SPECPARTS + RPM_EC=0 ++ jobs -p + exit 0 Executing(%prep): /bin/sh -e /var/tmp/rpm-tmp.tz75Ox + umask 022 + cd /builddir/build/BUILD/llama-cpp-b4094-build + cd /builddir/build/BUILD/llama-cpp-b4094-build + rm -rf llama.cpp-b4094 + /usr/lib/rpm/rpmuncompress -x /builddir/build/SOURCES/llama.cpp-b4094.tar.gz + STATUS=0 + '[' 0 -ne 0 ']' + cd llama.cpp-b4094 + /usr/bin/chmod -Rf a+rX,u+w,g-w,o-w . + sed -i -e 's/POSITION_INDEPENDENT_CODE ON/POSITION_INDEPENDENT_CODE ON SOVERSION b4094/' src/CMakeLists.txt + sed -i -e 's/POSITION_INDEPENDENT_CODE ON/POSITION_INDEPENDENT_CODE ON SOVERSION b4094/' ggml/src/CMakeLists.txt + rm -rf exmples/llma.android + find . -name .gitignore -exec rm -rf '{}' ';' + RPM_EC=0 ++ jobs -p + exit 0 Executing(%build): /bin/sh -e /var/tmp/rpm-tmp.TIwm1k + umask 022 + cd /builddir/build/BUILD/llama-cpp-b4094-build + CFLAGS='-O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer ' + export CFLAGS + CXXFLAGS='-O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer' + export CXXFLAGS + FFLAGS='-O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -I/usr/lib64/gfortran/modules ' + export FFLAGS + FCFLAGS='-O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -I/usr/lib64/gfortran/modules ' + export FCFLAGS + VALAFLAGS=-g + export VALAFLAGS + RUSTFLAGS='-Copt-level=3 -Cdebuginfo=2 -Ccodegen-units=1 -Cstrip=none -Cforce-frame-pointers=yes -Clink-arg=-specs=/usr/lib/rpm/redhat/redhat-package-notes --cap-lints=warn' + export RUSTFLAGS + LDFLAGS='-Wl,-z,relro -Wl,--as-needed -Wl,-z,pack-relative-relocs -Wl,-z,now -Wl,-z,now -Wl,--build-id=sha1 -specs=/usr/lib/rpm/redhat/redhat-package-notes ' + export LDFLAGS + LT_SYS_LIBRARY_PATH=/usr/lib64: + export LT_SYS_LIBRARY_PATH + CC=hipcc + export CC + CXX=hipcc + export CXX + cd llama.cpp-b4094 + module load rocm/default + local _mlredir=0 + '[' -n '' ']' + case " $@ " in + '[' 0 -eq 0 ']' + _module_raw load rocm/default ++ /usr/bin/tclsh /usr/share/Modules/libexec/modulecmd.tcl bash load rocm/default + eval '__MODULES_LMCONFLICT=rocm/default\&rocm; export __MODULES_LMCONFLICT; ROCM_BIN=/usr/bin; export ROCM_BIN; _LMFILES_=/usr/share/modulefiles/rocm/default; export _LMFILES_; LOADEDMODULES=rocm/default; export LOADEDMODULES; ROCM_GPUS=gfx900\;gfx906:xnack-\;gfx908:xnack-\;gfx90a:xnack+\;gfx90a:xnack-\;gfx942\;gfx1010\;gfx1012\;gfx1030\;gfx1031\;gfx1035\;gfx1100\;gfx1101\;gfx1102\;gfx1103\;gfx1151; export ROCM_GPUS; ROCM_LIB=/usr/lib64; export ROCM_LIB; test 0;' ++ __MODULES_LMCONFLICT='rocm/default&rocm' ++ export __MODULES_LMCONFLICT ++ ROCM_BIN=/usr/bin ++ export ROCM_BIN ++ _LMFILES_=/usr/share/modulefiles/rocm/default ++ export _LMFILES_ ++ LOADEDMODULES=rocm/default ++ export LOADEDMODULES ++ ROCM_GPUS='gfx900;gfx906:xnack-;gfx908:xnack-;gfx90a:xnack+;gfx90a:xnack-;gfx942;gfx1010;gfx1012;gfx1030;gfx1031;gfx1035;gfx1100;gfx1101;gfx1102;gfx1103;gfx1151' ++ export ROCM_GPUS ++ ROCM_LIB=/usr/lib64 ++ export ROCM_LIB ++ test 0 + _mlstatus=0 + return 0 + CFLAGS='-O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer ' + export CFLAGS + CXXFLAGS='-O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer' + export CXXFLAGS + FFLAGS='-O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -I/usr/lib64/gfortran/modules ' + export FFLAGS + FCFLAGS='-O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -I/usr/lib64/gfortran/modules ' + export FCFLAGS + VALAFLAGS=-g + export VALAFLAGS + RUSTFLAGS='-Copt-level=3 -Cdebuginfo=2 -Ccodegen-units=1 -Cstrip=none -Cforce-frame-pointers=yes -Clink-arg=-specs=/usr/lib/rpm/redhat/redhat-package-notes --cap-lints=warn' + export RUSTFLAGS + LDFLAGS='-Wl,-z,relro -Wl,--as-needed -Wl,-z,pack-relative-relocs -Wl,-z,now -Wl,-z,now -Wl,--build-id=sha1 -specs=/usr/lib/rpm/redhat/redhat-package-notes ' + export LDFLAGS + LT_SYS_LIBRARY_PATH=/usr/lib64: + export LT_SYS_LIBRARY_PATH + CC=hipcc + export CC + CXX=hipcc + export CXX + /usr/bin/cmake -S . -B redhat-linux-build -DCMAKE_C_FLAGS_RELEASE:STRING=-DNDEBUG -DCMAKE_CXX_FLAGS_RELEASE:STRING=-DNDEBUG -DCMAKE_Fortran_FLAGS_RELEASE:STRING=-DNDEBUG -DCMAKE_VERBOSE_MAKEFILE:BOOL=ON -DCMAKE_INSTALL_DO_STRIP:BOOL=OFF -DCMAKE_INSTALL_PREFIX:PATH=/usr -DINCLUDE_INSTALL_DIR:PATH=/usr/include -DLIB_INSTALL_DIR:PATH=/usr/lib64 -DSYSCONF_INSTALL_DIR:PATH=/etc -DSHARE_INSTALL_PREFIX:PATH=/usr/share -DLIB_SUFFIX=64 -DBUILD_SHARED_LIBS:BOOL=ON -DCMAKE_INSTALL_LIBDIR=lib64 -DCMAKE_SKIP_RPATH=ON -DGGML_AVX=OFF -DGGML_AVX2=OFF -DGGML_AVX512=OFF -DGGML_AVX512_VBMI=OFF -DGGML_AVX512_VNNI=OFF -DGGML_FMA=OFF -DGGML_F16C=OFF -DGGML_HIP=ON '-DAMDGPU_TARGETS=gfx900;gfx906:xnack-;gfx908:xnack-;gfx90a:xnack+;gfx90a:xnack-;gfx942;gfx1010;gfx1012;gfx1030;gfx1031;gfx1035;gfx1100;gfx1101;gfx1102;gfx1103;gfx1151' -DLLAMA_BUILD_EXAMPLES=OFF -DLLAMA_BUILD_TESTS=OFF -- The C compiler identification is Clang 18.0.0 -- The CXX compiler identification is Clang 18.0.0 -- Detecting C compiler ABI info -- Detecting C compiler ABI info - done -- Check for working C compiler: /usr/bin/hipcc - skipped -- Detecting C compile features -- Detecting C compile features - done -- Detecting CXX compiler ABI info -- Detecting CXX compiler ABI info - done -- Check for working CXX compiler: /usr/bin/hipcc - skipped -- Detecting CXX compile features -- Detecting CXX compile features - done -- Could NOT find Git (missing: GIT_EXECUTABLE) -- Could NOT find Git (missing: GIT_EXECUTABLE) CMake Warning at cmake/build-info.cmake:14 (message): Git not found. Build info will not be accurate. Call Stack (most recent call first): CMakeLists.txt:77 (include) sh: line 1: /usr/bin/rocm_agent_enumerator: No such file or directory sh: line 1: /usr/bin/rocm_agent_enumerator: No such file or directory -- Performing Test CMAKE_HAVE_LIBC_PTHREAD -- Performing Test CMAKE_HAVE_LIBC_PTHREAD - Success -- Found Threads: TRUE -- Warning: ccache not found - consider installing it for faster compilation or disable this warning with GGML_CCACHE=OFF -- CMAKE_SYSTEM_PROCESSOR: x86_64 -- Could NOT find OpenMP_C (missing: OpenMP_C_FLAGS OpenMP_C_LIB_NAMES) -- Could NOT find OpenMP_CXX (missing: OpenMP_CXX_FLAGS OpenMP_CXX_LIB_NAMES) -- Could NOT find OpenMP (missing: OpenMP_C_FOUND OpenMP_CXX_FOUND) CMake Warning at ggml/src/ggml-cpu/CMakeLists.txt:43 (message): OpenMP not found -- Using llamafile -- x86 detected -- Using runtime weight conversion of Q4_0 to Q4_0_x_x to enable optimized GEMM/GEMV kernels -- Including CPU backend CMake Warning at ggml/src/ggml-amx/CMakeLists.txt:106 (message): AMX requires x86 and gcc version > 11.0. Turning off GGML_AMX. CMake Warning at ggml/src/ggml-hip/CMakeLists.txt:27 (message): Setting hipcc as the C++ compiler is legacy behavior. Prefer setting the HIP compiler directly. See README for details. -- Performing Test HIP_CLANG_SUPPORTS_PARALLEL_JOBS -- Performing Test HIP_CLANG_SUPPORTS_PARALLEL_JOBS - Success -- HIP and hipBLAS found -- Including HIP backend CMake Warning at common/CMakeLists.txt:30 (message): Git repository not found; to enable automatic generation of build info, make sure Git is installed and the project is a Git repository. -- Configuring done (7.5s) -- Generating done (0.0s) CMake Warning: Manually-specified variables were not used by the project: CMAKE_Fortran_FLAGS_RELEASE CMAKE_INSTALL_DO_STRIP INCLUDE_INSTALL_DIR LIB_INSTALL_DIR LIB_SUFFIX SHARE_INSTALL_PREFIX SYSCONF_INSTALL_DIR -- Build files have been written to: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build + /usr/bin/cmake --build redhat-linux-build -j2 --verbose Change Dir: '/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build' Run Build Command(s): /usr/bin/cmake -E env VERBOSE=1 /usr/bin/gmake -f Makefile -j2 /usr/bin/cmake -S/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094 -B/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build --check-build-system CMakeFiles/Makefile.cmake 0 /usr/bin/cmake -E cmake_progress_start /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/CMakeFiles /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build//CMakeFiles/progress.marks /usr/bin/gmake -f CMakeFiles/Makefile2 all gmake[1]: Entering directory '/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build' /usr/bin/gmake -f ggml/src/CMakeFiles/ggml-base.dir/build.make ggml/src/CMakeFiles/ggml-base.dir/depend /usr/bin/gmake -f common/CMakeFiles/build_info.dir/build.make common/CMakeFiles/build_info.dir/depend gmake[2]: Entering directory '/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build' gmake[2]: Entering directory '/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build' cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build && /usr/bin/cmake -E cmake_depends "Unix Makefiles" /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094 /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/CMakeFiles/ggml-base.dir/DependInfo.cmake "--color=" [ 0%] Generating build details from Git cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094 && /usr/bin/cmake -DMSVC= -DCMAKE_C_COMPILER_VERSION=18.0.0 -DCMAKE_C_COMPILER_ID=Clang -DCMAKE_VS_PLATFORM_NAME= -DCMAKE_C_COMPILER=/usr/bin/hipcc -P /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/common/cmake/build-info-gen-cpp.cmake gmake[2]: Leaving directory '/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build' /usr/bin/gmake -f ggml/src/CMakeFiles/ggml-base.dir/build.make ggml/src/CMakeFiles/ggml-base.dir/build gmake[2]: Entering directory '/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build' -- Could NOT find Git (missing: GIT_EXECUTABLE) Hint: The project() command has not yet been called. It sets up system-specific search paths. CMake Warning at cmake/build-info.cmake:14 (message): Git not found. Build info will not be accurate. Call Stack (most recent call first): common/cmake/build-info-gen-cpp.cmake:1 (include) [ 1%] Building C object ggml/src/CMakeFiles/ggml-base.dir/ggml.c.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src && /usr/bin/hipcc -DGGML_BUILD -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -Dggml_base_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu11 -fPIC -Wshadow -Wstrict-prototypes -Wpointer-arith -Wmissing-prototypes -Werror=implicit-int -Werror=implicit-function-declaration -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wdouble-promotion -MD -MT ggml/src/CMakeFiles/ggml-base.dir/ggml.c.o -MF CMakeFiles/ggml-base.dir/ggml.c.o.d -o CMakeFiles/ggml-base.dir/ggml.c.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml.c sh: line 1: /usr/bin/rocm_agent_enumerator: No such file or directory sh: line 1: /usr/bin/rocm_agent_enumerator: No such file or directory sh: line 1: /usr/bin/rocm_agent_enumerator: No such file or directory cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build && /usr/bin/cmake -E cmake_depends "Unix Makefiles" /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094 /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/common /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/common /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/common/CMakeFiles/build_info.dir/DependInfo.cmake "--color=" gmake[2]: Leaving directory '/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build' /usr/bin/gmake -f common/CMakeFiles/build_info.dir/build.make common/CMakeFiles/build_info.dir/build gmake[2]: Entering directory '/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build' [ 2%] Building CXX object common/CMakeFiles/build_info.dir/build-info.cpp.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/common && /usr/bin/hipcc -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -fPIC -MD -MT common/CMakeFiles/build_info.dir/build-info.cpp.o -MF CMakeFiles/build_info.dir/build-info.cpp.o.d -o CMakeFiles/build_info.dir/build-info.cpp.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/common/build-info.cpp sh: line 1: /usr/bin/rocm_agent_enumerator: No such file or directory [ 3%] Building C object ggml/src/CMakeFiles/ggml-base.dir/ggml-alloc.c.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src && /usr/bin/hipcc -DGGML_BUILD -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -Dggml_base_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu11 -fPIC -Wshadow -Wstrict-prototypes -Wpointer-arith -Wmissing-prototypes -Werror=implicit-int -Werror=implicit-function-declaration -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wdouble-promotion -MD -MT ggml/src/CMakeFiles/ggml-base.dir/ggml-alloc.c.o -MF CMakeFiles/ggml-base.dir/ggml-alloc.c.o.d -o CMakeFiles/ggml-base.dir/ggml-alloc.c.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-alloc.c sh: line 1: /usr/bin/rocm_agent_enumerator: No such file or directory [ 4%] Building CXX object ggml/src/CMakeFiles/ggml-base.dir/ggml-backend.cpp.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src && /usr/bin/hipcc -DGGML_BUILD -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -Dggml_base_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -MD -MT ggml/src/CMakeFiles/ggml-base.dir/ggml-backend.cpp.o -MF CMakeFiles/ggml-base.dir/ggml-backend.cpp.o.d -o CMakeFiles/ggml-base.dir/ggml-backend.cpp.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-backend.cpp sh: line 1: /usr/bin/rocm_agent_enumerator: No such file or directory gmake[2]: Leaving directory '/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build' [ 4%] Built target build_info [ 5%] Building CXX object ggml/src/CMakeFiles/ggml-base.dir/ggml-threading.cpp.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src && /usr/bin/hipcc -DGGML_BUILD -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -Dggml_base_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -MD -MT ggml/src/CMakeFiles/ggml-base.dir/ggml-threading.cpp.o -MF CMakeFiles/ggml-base.dir/ggml-threading.cpp.o.d -o CMakeFiles/ggml-base.dir/ggml-threading.cpp.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-threading.cpp sh: line 1: /usr/bin/rocm_agent_enumerator: No such file or directory [ 6%] Building C object ggml/src/CMakeFiles/ggml-base.dir/ggml-quants.c.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src && /usr/bin/hipcc -DGGML_BUILD -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -Dggml_base_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu11 -fPIC -Wshadow -Wstrict-prototypes -Wpointer-arith -Wmissing-prototypes -Werror=implicit-int -Werror=implicit-function-declaration -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wdouble-promotion -MD -MT ggml/src/CMakeFiles/ggml-base.dir/ggml-quants.c.o -MF CMakeFiles/ggml-base.dir/ggml-quants.c.o.d -o CMakeFiles/ggml-base.dir/ggml-quants.c.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-quants.c sh: line 1: /usr/bin/rocm_agent_enumerator: No such file or directory [ 7%] Building C object ggml/src/CMakeFiles/ggml-base.dir/ggml-aarch64.c.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src && /usr/bin/hipcc -DGGML_BUILD -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -Dggml_base_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu11 -fPIC -Wshadow -Wstrict-prototypes -Wpointer-arith -Wmissing-prototypes -Werror=implicit-int -Werror=implicit-function-declaration -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wdouble-promotion -MD -MT ggml/src/CMakeFiles/ggml-base.dir/ggml-aarch64.c.o -MF CMakeFiles/ggml-base.dir/ggml-aarch64.c.o.d -o CMakeFiles/ggml-base.dir/ggml-aarch64.c.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-aarch64.c sh: line 1: /usr/bin/rocm_agent_enumerator: No such file or directory [ 8%] Linking CXX shared library libggml-base.so cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src && /usr/bin/cmake -E cmake_link_script CMakeFiles/ggml-base.dir/link.txt --verbose=1 sh: line 1: /usr/bin/rocm_agent_enumerator: No such file or directory clang++: warning: argument unused during compilation: '-Xarch_host -fstack-protector-strong' [-Wunused-command-line-argument] clang++: warning: argument unused during compilation: '-Xarch_host -fcf-protection' [-Wunused-command-line-argument] clang++: warning: argument unused during compilation: '-specs=/usr/lib/rpm/redhat/redhat-package-notes' [-Wunused-command-line-argument] /usr/bin/hipcc -fPIC -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -Xlinker --dependency-file=CMakeFiles/ggml-base.dir/link.d -Wl,-z,relro -Wl,--as-needed -Wl,-z,pack-relative-relocs -Wl,-z,now -Wl,-z,now -Wl,--build-id=sha1 -specs=/usr/lib/rpm/redhat/redhat-package-notes -shared -Wl,-soname,libggml-base.so.b4094 -o libggml-base.so.b4094 "CMakeFiles/ggml-base.dir/ggml.c.o" "CMakeFiles/ggml-base.dir/ggml-alloc.c.o" "CMakeFiles/ggml-base.dir/ggml-backend.cpp.o" "CMakeFiles/ggml-base.dir/ggml-threading.cpp.o" "CMakeFiles/ggml-base.dir/ggml-quants.c.o" "CMakeFiles/ggml-base.dir/ggml-aarch64.c.o" -lm cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src && /usr/bin/cmake -E cmake_symlink_library libggml-base.so.b4094 libggml-base.so.b4094 libggml-base.so gmake[2]: Leaving directory '/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build' [ 8%] Built target ggml-base /usr/bin/gmake -f ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/build.make ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/depend /usr/bin/gmake -f ggml/src/ggml-cpu/CMakeFiles/ggml-cpu.dir/build.make ggml/src/ggml-cpu/CMakeFiles/ggml-cpu.dir/depend gmake[2]: Entering directory '/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build' cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build && /usr/bin/cmake -E cmake_depends "Unix Makefiles" /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094 /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cpu /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-cpu /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-cpu/CMakeFiles/ggml-cpu.dir/DependInfo.cmake "--color=" gmake[2]: Entering directory '/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build' cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build && /usr/bin/cmake -E cmake_depends "Unix Makefiles" /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094 /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/DependInfo.cmake "--color=" gmake[2]: Leaving directory '/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build' /usr/bin/gmake -f ggml/src/ggml-cpu/CMakeFiles/ggml-cpu.dir/build.make ggml/src/ggml-cpu/CMakeFiles/ggml-cpu.dir/build gmake[2]: Entering directory '/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build' gmake[2]: Leaving directory '/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build' /usr/bin/gmake -f ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/build.make ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/build gmake[2]: Entering directory '/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build' [ 9%] Building C object ggml/src/ggml-cpu/CMakeFiles/ggml-cpu.dir/ggml-cpu.c.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-cpu && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_CPU_AARCH64 -DGGML_USE_LLAMAFILE -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -Dggml_cpu_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cpu/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cpu/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu11 -fPIC -Wshadow -Wstrict-prototypes -Wpointer-arith -Wmissing-prototypes -Werror=implicit-int -Werror=implicit-function-declaration -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wdouble-promotion -march=native -MD -MT ggml/src/ggml-cpu/CMakeFiles/ggml-cpu.dir/ggml-cpu.c.o -MF CMakeFiles/ggml-cpu.dir/ggml-cpu.c.o.d -o CMakeFiles/ggml-cpu.dir/ggml-cpu.c.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cpu/ggml-cpu.c [ 10%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/acc.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/acc.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/acc.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/acc.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/acc.cu sh: line 1: /usr/bin/rocm_agent_enumerator: No such file or directory In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/acc.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/acc.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/acc.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/acc.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1012. [ 11%] Building CXX object ggml/src/ggml-cpu/CMakeFiles/ggml-cpu.dir/ggml-cpu.cpp.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-cpu && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_CPU_AARCH64 -DGGML_USE_LLAMAFILE -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -Dggml_cpu_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cpu/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cpu/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -march=native -MD -MT ggml/src/ggml-cpu/CMakeFiles/ggml-cpu.dir/ggml-cpu.cpp.o -MF CMakeFiles/ggml-cpu.dir/ggml-cpu.cpp.o.d -o CMakeFiles/ggml-cpu.dir/ggml-cpu.cpp.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cpu/ggml-cpu.cpp sh: line 1: /usr/bin/rocm_agent_enumerator: No such file or directory In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/acc.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/acc.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/acc.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/acc.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1031. [ 12%] Building C object ggml/src/ggml-cpu/CMakeFiles/ggml-cpu.dir/ggml-cpu-aarch64.c.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-cpu && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_CPU_AARCH64 -DGGML_USE_LLAMAFILE -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -Dggml_cpu_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cpu/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cpu/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu11 -fPIC -Wshadow -Wstrict-prototypes -Wpointer-arith -Wmissing-prototypes -Werror=implicit-int -Werror=implicit-function-declaration -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wdouble-promotion -march=native -MD -MT ggml/src/ggml-cpu/CMakeFiles/ggml-cpu.dir/ggml-cpu-aarch64.c.o -MF CMakeFiles/ggml-cpu.dir/ggml-cpu-aarch64.c.o.d -o CMakeFiles/ggml-cpu.dir/ggml-cpu-aarch64.c.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cpu/ggml-cpu-aarch64.c sh: line 1: /usr/bin/rocm_agent_enumerator: No such file or directory In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/acc.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/acc.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ [ 13%] Building C object ggml/src/ggml-cpu/CMakeFiles/ggml-cpu.dir/ggml-cpu-quants.c.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-cpu && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_CPU_AARCH64 -DGGML_USE_LLAMAFILE -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -Dggml_cpu_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cpu/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cpu/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu11 -fPIC -Wshadow -Wstrict-prototypes -Wpointer-arith -Wmissing-prototypes -Werror=implicit-int -Werror=implicit-function-declaration -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wdouble-promotion -march=native -MD -MT ggml/src/ggml-cpu/CMakeFiles/ggml-cpu.dir/ggml-cpu-quants.c.o -MF CMakeFiles/ggml-cpu.dir/ggml-cpu-quants.c.o.d -o CMakeFiles/ggml-cpu.dir/ggml-cpu-quants.c.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cpu/ggml-cpu-quants.c sh: line 1: /usr/bin/rocm_agent_enumerator: No such file or directory 6 warnings generated when compiling for gfx1035. [ 14%] Building CXX object ggml/src/ggml-cpu/CMakeFiles/ggml-cpu.dir/llamafile/sgemm.cpp.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-cpu && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_CPU_AARCH64 -DGGML_USE_LLAMAFILE -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -Dggml_cpu_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cpu/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cpu/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -march=native -MD -MT ggml/src/ggml-cpu/CMakeFiles/ggml-cpu.dir/llamafile/sgemm.cpp.o -MF CMakeFiles/ggml-cpu.dir/llamafile/sgemm.cpp.o.d -o CMakeFiles/ggml-cpu.dir/llamafile/sgemm.cpp.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cpu/llamafile/sgemm.cpp sh: line 1: /usr/bin/rocm_agent_enumerator: No such file or directory In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/acc.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/acc.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/acc.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/acc.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/acc.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/acc.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ [ 15%] Linking CXX shared library libggml-cpu.so cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-cpu && /usr/bin/cmake -E cmake_link_script CMakeFiles/ggml-cpu.dir/link.txt --verbose=1 sh: line 1: /usr/bin/rocm_agent_enumerator: No such file or directory clang++: warning: argument unused during compilation: '-Xarch_host -fstack-protector-strong' [-Wunused-command-line-argument] clang++: warning: argument unused during compilation: '-Xarch_host -fcf-protection' [-Wunused-command-line-argument] clang++: warning: argument unused during compilation: '-specs=/usr/lib/rpm/redhat/redhat-package-notes' [-Wunused-command-line-argument] 6 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/acc.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/acc.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/acc.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/acc.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/acc.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/acc.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/acc.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/acc.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx906. /usr/bin/hipcc -fPIC -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -Xlinker --dependency-file=CMakeFiles/ggml-cpu.dir/link.d -Wl,-z,relro -Wl,--as-needed -Wl,-z,pack-relative-relocs -Wl,-z,now -Wl,-z,now -Wl,--build-id=sha1 -specs=/usr/lib/rpm/redhat/redhat-package-notes -shared -Wl,-soname,libggml-cpu.so -o libggml-cpu.so "CMakeFiles/ggml-cpu.dir/ggml-cpu.c.o" "CMakeFiles/ggml-cpu.dir/ggml-cpu.cpp.o" "CMakeFiles/ggml-cpu.dir/ggml-cpu-aarch64.c.o" "CMakeFiles/ggml-cpu.dir/ggml-cpu-quants.c.o" "CMakeFiles/ggml-cpu.dir/llamafile/sgemm.cpp.o" ../libggml-base.so.b4094 gmake[2]: Leaving directory '/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build' [ 15%] Built target ggml-cpu [ 16%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/arange.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/arange.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/arange.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/arange.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/arange.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/acc.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/acc.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/arange.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/arange.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/acc.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/acc.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/arange.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/arange.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/acc.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/acc.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/arange.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/arange.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/acc.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/acc.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/arange.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/arange.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/acc.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/acc.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for host. [ 17%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/argmax.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/argmax.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/argmax.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/argmax.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argmax.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/arange.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/arange.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argmax.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/arange.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/arange.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1010. 6 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argmax.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/arange.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/arange.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1012. 6 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argmax.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/arange.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/arange.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1102. 6 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/arange.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/arange.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argmax.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1103. 6 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/arange.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/arange.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argmax.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1151. 6 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/arange.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/arange.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argmax.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/arange.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/arange.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1100. 6 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argmax.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/arange.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/arange.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1101. 6 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argmax.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/arange.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/arange.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx90a. 6 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/arange.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/arange.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argmax.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx90a. 6 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/arange.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/arange.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argmax.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/arange.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/arange.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for host. [ 18%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/argsort.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/argsort.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/argsort.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/argsort.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argsort.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argmax.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argsort.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argsort.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx900. 6 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argmax.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argsort.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argsort.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx906. 6 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argmax.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argsort.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argsort.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1030. 6 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argsort.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argsort.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argmax.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1031. 6 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argsort.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argsort.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argmax.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1035. 6 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argsort.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argsort.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argmax.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argsort.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argsort.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argmax.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for host. [ 19%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/binbcast.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/binbcast.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/binbcast.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/binbcast.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argsort.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argsort.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cu:353:11: warning: 'break' will never be executed [-Wunreachable-code-break] 353 | } break; | ^~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argsort.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argsort.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1103. 7 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argsort.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argsort.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1151. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cu:353:11: warning: 'break' will never be executed [-Wunreachable-code-break] 353 | } break; | ^~~~~ 7 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argsort.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argsort.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cu:353:11: warning: 'break' will never be executed [-Wunreachable-code-break] 353 | } break; | ^~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argsort.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argsort.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 7 warnings generated when compiling for gfx1030. 6 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argsort.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argsort.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cu:353:11: warning: 'break' will never be executed [-Wunreachable-code-break] 353 | } break; | ^~~~~ 6 warnings generated when compiling for gfx908. 7 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argsort.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argsort.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cu:353:11: warning: 'break' will never be executed [-Wunreachable-code-break] 353 | } break; | ^~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argsort.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argsort.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 7 warnings generated when compiling for gfx1035. 6 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argsort.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argsort.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cu:353:11: warning: 'break' will never be executed [-Wunreachable-code-break] 353 | } break; | ^~~~~ 6 warnings generated when compiling for gfx942. 7 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argsort.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/argsort.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for host. [ 20%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/clamp.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/clamp.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/clamp.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/clamp.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/clamp.cu /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cu:353:11: warning: 'break' will never be executed [-Wunreachable-code-break] 353 | } break; | ^~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/clamp.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/clamp.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 7 warnings generated when compiling for gfx1101. 6 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/clamp.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/clamp.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cu:353:11: warning: 'break' will never be executed [-Wunreachable-code-break] 353 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1012. 7 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/clamp.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/clamp.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1030. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cu:353:11: warning: 'break' will never be executed [-Wunreachable-code-break] 353 | } break; | ^~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/clamp.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/clamp.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 7 warnings generated when compiling for gfx1103. 6 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/clamp.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/clamp.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cu:353:11: warning: 'break' will never be executed [-Wunreachable-code-break] 353 | } break; | ^~~~~ 6 warnings generated when compiling for gfx1035. 7 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/clamp.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/clamp.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1100. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cu:353:11: warning: 'break' will never be executed [-Wunreachable-code-break] 353 | } break; | ^~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/clamp.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/clamp.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 7 warnings generated when compiling for gfx900. 6 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/clamp.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/clamp.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cu:353:11: warning: 'break' will never be executed [-Wunreachable-code-break] 353 | } break; | ^~~~~ 6 warnings generated when compiling for gfx1102. 7 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/clamp.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/clamp.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1103. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cu:353:11: warning: 'break' will never be executed [-Wunreachable-code-break] 353 | } break; | ^~~~~ 7 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/clamp.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/clamp.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cu:353:11: warning: 'break' will never be executed [-Wunreachable-code-break] 353 | } break; | ^~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/clamp.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/clamp.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx900. 7 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/clamp.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/clamp.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx906. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cu:353:11: warning: 'break' will never be executed [-Wunreachable-code-break] 353 | } break; | ^~~~~ 7 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/clamp.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/clamp.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cu:353:11: warning: 'break' will never be executed [-Wunreachable-code-break] 353 | } break; | ^~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/clamp.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/clamp.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx90a. 7 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/clamp.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/clamp.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/binbcast.cu:353:11: warning: 'break' will never be executed [-Wunreachable-code-break] 353 | } break; | ^~~~~ 6 warnings generated when compiling for gfx90a. 7 warnings generated when compiling for host. [ 21%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/concat.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/concat.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/concat.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/concat.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/clamp.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/clamp.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:41:20: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 41 | if (blockIdx.y < ne01) { // src0 | ~~~~~~~~~~ ^ ~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:67:20: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 67 | if (blockIdx.z < ne02) { // src0 | ~~~~~~~~~~ ^ ~~~~ 8 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/clamp.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/clamp.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for host. [ 22%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/conv-transpose-1d.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/conv-transpose-1d.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/conv-transpose-1d.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/conv-transpose-1d.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:41:20: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 41 | if (blockIdx.y < ne01) { // src0 | ~~~~~~~~~~ ^ ~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:67:20: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 67 | if (blockIdx.z < ne02) { // src0 | ~~~~~~~~~~ ^ ~~~~ 8 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:4:33: warning: unused parameter 'p0' [-Wunused-parameter] 4 | const int s0, const int p0, const int d0, const int output_size, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:4:47: warning: unused parameter 'd0' [-Wunused-parameter] 4 | const int s0, const int p0, const int d0, const int output_size, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:5:79: warning: unused parameter 'src0_ne3' [-Wunused-parameter] 5 | const int src0_ne0, const int src0_ne1, const int src0_ne2, const int src0_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:39: warning: unused parameter 'src1_ne1' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:59: warning: unused parameter 'src1_ne2' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:79: warning: unused parameter 'src1_ne3' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:38: warning: unused parameter 'dst_ne1' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:57: warning: unused parameter 'dst_ne2' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:76: warning: unused parameter 'dst_ne3' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:78:19: warning: unused variable 'kernel_size' [-Wunused-variable] 78 | const int64_t kernel_size = ggml_nelements(src0); | ^~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:79:19: warning: unused variable 'input_size' [-Wunused-variable] 79 | const int64_t input_size = ggml_nelements(src1); | ^~~~~~~~~~ 17 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:41:20: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 41 | if (blockIdx.y < ne01) { // src0 | ~~~~~~~~~~ ^ ~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:67:20: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 67 | if (blockIdx.z < ne02) { // src0 | ~~~~~~~~~~ ^ ~~~~ 8 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:4:33: warning: unused parameter 'p0' [-Wunused-parameter] 4 | const int s0, const int p0, const int d0, const int output_size, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:4:47: warning: unused parameter 'd0' [-Wunused-parameter] 4 | const int s0, const int p0, const int d0, const int output_size, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:5:79: warning: unused parameter 'src0_ne3' [-Wunused-parameter] 5 | const int src0_ne0, const int src0_ne1, const int src0_ne2, const int src0_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:39: warning: unused parameter 'src1_ne1' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:59: warning: unused parameter 'src1_ne2' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:79: warning: unused parameter 'src1_ne3' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:38: warning: unused parameter 'dst_ne1' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:57: warning: unused parameter 'dst_ne2' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:76: warning: unused parameter 'dst_ne3' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:78:19: warning: unused variable 'kernel_size' [-Wunused-variable] 78 | const int64_t kernel_size = ggml_nelements(src0); | ^~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:79:19: warning: unused variable 'input_size' [-Wunused-variable] 79 | const int64_t input_size = ggml_nelements(src1); | ^~~~~~~~~~ 17 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:41:20: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 41 | if (blockIdx.y < ne01) { // src0 | ~~~~~~~~~~ ^ ~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:67:20: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 67 | if (blockIdx.z < ne02) { // src0 | ~~~~~~~~~~ ^ ~~~~ 8 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:4:33: warning: unused parameter 'p0' [-Wunused-parameter] 4 | const int s0, const int p0, const int d0, const int output_size, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:4:47: warning: unused parameter 'd0' [-Wunused-parameter] 4 | const int s0, const int p0, const int d0, const int output_size, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:5:79: warning: unused parameter 'src0_ne3' [-Wunused-parameter] 5 | const int src0_ne0, const int src0_ne1, const int src0_ne2, const int src0_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:39: warning: unused parameter 'src1_ne1' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:59: warning: unused parameter 'src1_ne2' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:79: warning: unused parameter 'src1_ne3' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:38: warning: unused parameter 'dst_ne1' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:57: warning: unused parameter 'dst_ne2' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:76: warning: unused parameter 'dst_ne3' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:78:19: warning: unused variable 'kernel_size' [-Wunused-variable] 78 | const int64_t kernel_size = ggml_nelements(src0); | ^~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:79:19: warning: unused variable 'input_size' [-Wunused-variable] 79 | const int64_t input_size = ggml_nelements(src1); | ^~~~~~~~~~ 17 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:41:20: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 41 | if (blockIdx.y < ne01) { // src0 | ~~~~~~~~~~ ^ ~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:67:20: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 67 | if (blockIdx.z < ne02) { // src0 | ~~~~~~~~~~ ^ ~~~~ 8 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:4:33: warning: unused parameter 'p0' [-Wunused-parameter] 4 | const int s0, const int p0, const int d0, const int output_size, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:4:47: warning: unused parameter 'd0' [-Wunused-parameter] 4 | const int s0, const int p0, const int d0, const int output_size, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:5:79: warning: unused parameter 'src0_ne3' [-Wunused-parameter] 5 | const int src0_ne0, const int src0_ne1, const int src0_ne2, const int src0_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:39: warning: unused parameter 'src1_ne1' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:59: warning: unused parameter 'src1_ne2' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:79: warning: unused parameter 'src1_ne3' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:38: warning: unused parameter 'dst_ne1' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:57: warning: unused parameter 'dst_ne2' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:76: warning: unused parameter 'dst_ne3' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:78:19: warning: unused variable 'kernel_size' [-Wunused-variable] 78 | const int64_t kernel_size = ggml_nelements(src0); | ^~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:79:19: warning: unused variable 'input_size' [-Wunused-variable] 79 | const int64_t input_size = ggml_nelements(src1); | ^~~~~~~~~~ 17 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:41:20: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 41 | if (blockIdx.y < ne01) { // src0 | ~~~~~~~~~~ ^ ~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:67:20: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 67 | if (blockIdx.z < ne02) { // src0 | ~~~~~~~~~~ ^ ~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 8 warnings generated when compiling for gfx1100. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:4:33: warning: unused parameter 'p0' [-Wunused-parameter] 4 | const int s0, const int p0, const int d0, const int output_size, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:4:47: warning: unused parameter 'd0' [-Wunused-parameter] 4 | const int s0, const int p0, const int d0, const int output_size, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:5:79: warning: unused parameter 'src0_ne3' [-Wunused-parameter] 5 | const int src0_ne0, const int src0_ne1, const int src0_ne2, const int src0_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:39: warning: unused parameter 'src1_ne1' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:59: warning: unused parameter 'src1_ne2' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:79: warning: unused parameter 'src1_ne3' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:38: warning: unused parameter 'dst_ne1' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:57: warning: unused parameter 'dst_ne2' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:76: warning: unused parameter 'dst_ne3' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:78:19: warning: unused variable 'kernel_size' [-Wunused-variable] 78 | const int64_t kernel_size = ggml_nelements(src0); | ^~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:79:19: warning: unused variable 'input_size' [-Wunused-variable] 79 | const int64_t input_size = ggml_nelements(src1); | ^~~~~~~~~~ 17 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:41:20: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 41 | if (blockIdx.y < ne01) { // src0 | ~~~~~~~~~~ ^ ~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:67:20: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 67 | if (blockIdx.z < ne02) { // src0 | ~~~~~~~~~~ ^ ~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 8 warnings generated when compiling for gfx1101. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:4:33: warning: unused parameter 'p0' [-Wunused-parameter] 4 | const int s0, const int p0, const int d0, const int output_size, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:4:47: warning: unused parameter 'd0' [-Wunused-parameter] 4 | const int s0, const int p0, const int d0, const int output_size, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:5:79: warning: unused parameter 'src0_ne3' [-Wunused-parameter] 5 | const int src0_ne0, const int src0_ne1, const int src0_ne2, const int src0_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:39: warning: unused parameter 'src1_ne1' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:59: warning: unused parameter 'src1_ne2' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:79: warning: unused parameter 'src1_ne3' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:38: warning: unused parameter 'dst_ne1' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:57: warning: unused parameter 'dst_ne2' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:76: warning: unused parameter 'dst_ne3' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:78:19: warning: unused variable 'kernel_size' [-Wunused-variable] 78 | const int64_t kernel_size = ggml_nelements(src0); | ^~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:79:19: warning: unused variable 'input_size' [-Wunused-variable] 79 | const int64_t input_size = ggml_nelements(src1); | ^~~~~~~~~~ 17 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:41:20: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 41 | if (blockIdx.y < ne01) { // src0 | ~~~~~~~~~~ ^ ~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:67:20: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 67 | if (blockIdx.z < ne02) { // src0 | ~~~~~~~~~~ ^ ~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 8 warnings generated when compiling for gfx1102. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:4:33: warning: unused parameter 'p0' [-Wunused-parameter] 4 | const int s0, const int p0, const int d0, const int output_size, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:4:47: warning: unused parameter 'd0' [-Wunused-parameter] 4 | const int s0, const int p0, const int d0, const int output_size, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:5:79: warning: unused parameter 'src0_ne3' [-Wunused-parameter] 5 | const int src0_ne0, const int src0_ne1, const int src0_ne2, const int src0_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:39: warning: unused parameter 'src1_ne1' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:59: warning: unused parameter 'src1_ne2' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:79: warning: unused parameter 'src1_ne3' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:38: warning: unused parameter 'dst_ne1' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:57: warning: unused parameter 'dst_ne2' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:76: warning: unused parameter 'dst_ne3' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:78:19: warning: unused variable 'kernel_size' [-Wunused-variable] 78 | const int64_t kernel_size = ggml_nelements(src0); | ^~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:79:19: warning: unused variable 'input_size' [-Wunused-variable] 79 | const int64_t input_size = ggml_nelements(src1); | ^~~~~~~~~~ 17 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:41:20: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 41 | if (blockIdx.y < ne01) { // src0 | ~~~~~~~~~~ ^ ~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:67:20: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 67 | if (blockIdx.z < ne02) { // src0 | ~~~~~~~~~~ ^ ~~~~ 8 warnings generated when compiling for gfx1103. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:4:33: warning: unused parameter 'p0' [-Wunused-parameter] 4 | const int s0, const int p0, const int d0, const int output_size, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:4:47: warning: unused parameter 'd0' [-Wunused-parameter] 4 | const int s0, const int p0, const int d0, const int output_size, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:5:79: warning: unused parameter 'src0_ne3' [-Wunused-parameter] 5 | const int src0_ne0, const int src0_ne1, const int src0_ne2, const int src0_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:39: warning: unused parameter 'src1_ne1' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:59: warning: unused parameter 'src1_ne2' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:79: warning: unused parameter 'src1_ne3' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:38: warning: unused parameter 'dst_ne1' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:57: warning: unused parameter 'dst_ne2' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:76: warning: unused parameter 'dst_ne3' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:78:19: warning: unused variable 'kernel_size' [-Wunused-variable] 78 | const int64_t kernel_size = ggml_nelements(src0); | ^~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:79:19: warning: unused variable 'input_size' [-Wunused-variable] 79 | const int64_t input_size = ggml_nelements(src1); | ^~~~~~~~~~ 17 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:41:20: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 41 | if (blockIdx.y < ne01) { // src0 | ~~~~~~~~~~ ^ ~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:67:20: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 67 | if (blockIdx.z < ne02) { // src0 | ~~~~~~~~~~ ^ ~~~~ 8 warnings generated when compiling for gfx1151. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:4:33: warning: unused parameter 'p0' [-Wunused-parameter] 4 | const int s0, const int p0, const int d0, const int output_size, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:4:47: warning: unused parameter 'd0' [-Wunused-parameter] 4 | const int s0, const int p0, const int d0, const int output_size, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:5:79: warning: unused parameter 'src0_ne3' [-Wunused-parameter] 5 | const int src0_ne0, const int src0_ne1, const int src0_ne2, const int src0_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:39: warning: unused parameter 'src1_ne1' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:59: warning: unused parameter 'src1_ne2' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:79: warning: unused parameter 'src1_ne3' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:38: warning: unused parameter 'dst_ne1' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:57: warning: unused parameter 'dst_ne2' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:76: warning: unused parameter 'dst_ne3' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:78:19: warning: unused variable 'kernel_size' [-Wunused-variable] 78 | const int64_t kernel_size = ggml_nelements(src0); | ^~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:79:19: warning: unused variable 'input_size' [-Wunused-variable] 79 | const int64_t input_size = ggml_nelements(src1); | ^~~~~~~~~~ 17 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:41:20: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 41 | if (blockIdx.y < ne01) { // src0 | ~~~~~~~~~~ ^ ~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:67:20: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 67 | if (blockIdx.z < ne02) { // src0 | ~~~~~~~~~~ ^ ~~~~ 8 warnings generated when compiling for gfx900. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:4:33: warning: unused parameter 'p0' [-Wunused-parameter] 4 | const int s0, const int p0, const int d0, const int output_size, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:4:47: warning: unused parameter 'd0' [-Wunused-parameter] 4 | const int s0, const int p0, const int d0, const int output_size, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:5:79: warning: unused parameter 'src0_ne3' [-Wunused-parameter] 5 | const int src0_ne0, const int src0_ne1, const int src0_ne2, const int src0_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:39: warning: unused parameter 'src1_ne1' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:59: warning: unused parameter 'src1_ne2' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:79: warning: unused parameter 'src1_ne3' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:38: warning: unused parameter 'dst_ne1' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:57: warning: unused parameter 'dst_ne2' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:76: warning: unused parameter 'dst_ne3' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:78:19: warning: unused variable 'kernel_size' [-Wunused-variable] 78 | const int64_t kernel_size = ggml_nelements(src0); | ^~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:79:19: warning: unused variable 'input_size' [-Wunused-variable] 79 | const int64_t input_size = ggml_nelements(src1); | ^~~~~~~~~~ 17 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:41:20: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 41 | if (blockIdx.y < ne01) { // src0 | ~~~~~~~~~~ ^ ~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:67:20: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 67 | if (blockIdx.z < ne02) { // src0 | ~~~~~~~~~~ ^ ~~~~ 8 warnings generated when compiling for gfx906. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:4:33: warning: unused parameter 'p0' [-Wunused-parameter] 4 | const int s0, const int p0, const int d0, const int output_size, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:4:47: warning: unused parameter 'd0' [-Wunused-parameter] 4 | const int s0, const int p0, const int d0, const int output_size, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:5:79: warning: unused parameter 'src0_ne3' [-Wunused-parameter] 5 | const int src0_ne0, const int src0_ne1, const int src0_ne2, const int src0_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:39: warning: unused parameter 'src1_ne1' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:59: warning: unused parameter 'src1_ne2' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:79: warning: unused parameter 'src1_ne3' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:38: warning: unused parameter 'dst_ne1' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:57: warning: unused parameter 'dst_ne2' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:76: warning: unused parameter 'dst_ne3' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:78:19: warning: unused variable 'kernel_size' [-Wunused-variable] 78 | const int64_t kernel_size = ggml_nelements(src0); | ^~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:79:19: warning: unused variable 'input_size' [-Wunused-variable] 79 | const int64_t input_size = ggml_nelements(src1); | ^~~~~~~~~~ 17 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:41:20: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 41 | if (blockIdx.y < ne01) { // src0 | ~~~~~~~~~~ ^ ~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:67:20: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 67 | if (blockIdx.z < ne02) { // src0 | ~~~~~~~~~~ ^ ~~~~ 8 warnings generated when compiling for gfx908. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:4:33: warning: unused parameter 'p0' [-Wunused-parameter] 4 | const int s0, const int p0, const int d0, const int output_size, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:4:47: warning: unused parameter 'd0' [-Wunused-parameter] 4 | const int s0, const int p0, const int d0, const int output_size, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:5:79: warning: unused parameter 'src0_ne3' [-Wunused-parameter] 5 | const int src0_ne0, const int src0_ne1, const int src0_ne2, const int src0_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:39: warning: unused parameter 'src1_ne1' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:59: warning: unused parameter 'src1_ne2' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:79: warning: unused parameter 'src1_ne3' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:38: warning: unused parameter 'dst_ne1' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:57: warning: unused parameter 'dst_ne2' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:76: warning: unused parameter 'dst_ne3' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:78:19: warning: unused variable 'kernel_size' [-Wunused-variable] 78 | const int64_t kernel_size = ggml_nelements(src0); | ^~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:79:19: warning: unused variable 'input_size' [-Wunused-variable] 79 | const int64_t input_size = ggml_nelements(src1); | ^~~~~~~~~~ 17 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:41:20: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 41 | if (blockIdx.y < ne01) { // src0 | ~~~~~~~~~~ ^ ~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:67:20: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 67 | if (blockIdx.z < ne02) { // src0 | ~~~~~~~~~~ ^ ~~~~ 8 warnings generated when compiling for gfx90a. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:4:33: warning: unused parameter 'p0' [-Wunused-parameter] 4 | const int s0, const int p0, const int d0, const int output_size, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:4:47: warning: unused parameter 'd0' [-Wunused-parameter] 4 | const int s0, const int p0, const int d0, const int output_size, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:5:79: warning: unused parameter 'src0_ne3' [-Wunused-parameter] 5 | const int src0_ne0, const int src0_ne1, const int src0_ne2, const int src0_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:39: warning: unused parameter 'src1_ne1' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:59: warning: unused parameter 'src1_ne2' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:79: warning: unused parameter 'src1_ne3' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:38: warning: unused parameter 'dst_ne1' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:57: warning: unused parameter 'dst_ne2' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:76: warning: unused parameter 'dst_ne3' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:78:19: warning: unused variable 'kernel_size' [-Wunused-variable] 78 | const int64_t kernel_size = ggml_nelements(src0); | ^~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:79:19: warning: unused variable 'input_size' [-Wunused-variable] 79 | const int64_t input_size = ggml_nelements(src1); | ^~~~~~~~~~ 17 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:41:20: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 41 | if (blockIdx.y < ne01) { // src0 | ~~~~~~~~~~ ^ ~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:67:20: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 67 | if (blockIdx.z < ne02) { // src0 | ~~~~~~~~~~ ^ ~~~~ 8 warnings generated when compiling for gfx90a. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:4:33: warning: unused parameter 'p0' [-Wunused-parameter] 4 | const int s0, const int p0, const int d0, const int output_size, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:4:47: warning: unused parameter 'd0' [-Wunused-parameter] 4 | const int s0, const int p0, const int d0, const int output_size, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:5:79: warning: unused parameter 'src0_ne3' [-Wunused-parameter] 5 | const int src0_ne0, const int src0_ne1, const int src0_ne2, const int src0_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:39: warning: unused parameter 'src1_ne1' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:59: warning: unused parameter 'src1_ne2' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:79: warning: unused parameter 'src1_ne3' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:38: warning: unused parameter 'dst_ne1' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:57: warning: unused parameter 'dst_ne2' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:76: warning: unused parameter 'dst_ne3' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:78:19: warning: unused variable 'kernel_size' [-Wunused-variable] 78 | const int64_t kernel_size = ggml_nelements(src0); | ^~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:79:19: warning: unused variable 'input_size' [-Wunused-variable] 79 | const int64_t input_size = ggml_nelements(src1); | ^~~~~~~~~~ 17 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:41:20: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 41 | if (blockIdx.y < ne01) { // src0 | ~~~~~~~~~~ ^ ~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:67:20: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 67 | if (blockIdx.z < ne02) { // src0 | ~~~~~~~~~~ ^ ~~~~ 8 warnings generated when compiling for gfx942. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:4:33: warning: unused parameter 'p0' [-Wunused-parameter] 4 | const int s0, const int p0, const int d0, const int output_size, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:4:47: warning: unused parameter 'd0' [-Wunused-parameter] 4 | const int s0, const int p0, const int d0, const int output_size, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:5:79: warning: unused parameter 'src0_ne3' [-Wunused-parameter] 5 | const int src0_ne0, const int src0_ne1, const int src0_ne2, const int src0_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:39: warning: unused parameter 'src1_ne1' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:59: warning: unused parameter 'src1_ne2' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:79: warning: unused parameter 'src1_ne3' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:38: warning: unused parameter 'dst_ne1' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:57: warning: unused parameter 'dst_ne2' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:76: warning: unused parameter 'dst_ne3' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:78:19: warning: unused variable 'kernel_size' [-Wunused-variable] 78 | const int64_t kernel_size = ggml_nelements(src0); | ^~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:79:19: warning: unused variable 'input_size' [-Wunused-variable] 79 | const int64_t input_size = ggml_nelements(src1); | ^~~~~~~~~~ 17 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:41:20: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 41 | if (blockIdx.y < ne01) { // src0 | ~~~~~~~~~~ ^ ~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/concat.cu:67:20: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 67 | if (blockIdx.z < ne02) { // src0 | ~~~~~~~~~~ ^ ~~~~ 8 warnings generated when compiling for host. [ 23%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/convert.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/convert.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/convert.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/convert.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:4:33: warning: unused parameter 'p0' [-Wunused-parameter] 4 | const int s0, const int p0, const int d0, const int output_size, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:4:47: warning: unused parameter 'd0' [-Wunused-parameter] 4 | const int s0, const int p0, const int d0, const int output_size, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:5:79: warning: unused parameter 'src0_ne3' [-Wunused-parameter] 5 | const int src0_ne0, const int src0_ne1, const int src0_ne2, const int src0_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:39: warning: unused parameter 'src1_ne1' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:59: warning: unused parameter 'src1_ne2' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:79: warning: unused parameter 'src1_ne3' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:38: warning: unused parameter 'dst_ne1' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:57: warning: unused parameter 'dst_ne2' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:76: warning: unused parameter 'dst_ne3' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:78:19: warning: unused variable 'kernel_size' [-Wunused-variable] 78 | const int64_t kernel_size = ggml_nelements(src0); | ^~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:79:19: warning: unused variable 'input_size' [-Wunused-variable] 79 | const int64_t input_size = ggml_nelements(src1); | ^~~~~~~~~~ 17 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *__restrict' to 'type-parameter-0-0 *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:467:9: note: in instantiation of function template specialization 'dequantize_block_q8_0_f16' requested here 467 | dequantize_block_q8_0_f16<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:470:9: note: in instantiation of function template specialization 'dequantize_block_q8_0_f16' requested here 470 | dequantize_block_q8_0_f16<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *' to 'float *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:588:5: note: in instantiation of function template specialization 'convert_unary' requested here 588 | convert_unary<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:635:20: note: in instantiation of function template specialization 'convert_unary_cuda' requested here 635 | return convert_unary_cuda; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *' to '__half *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:588:5: note: in instantiation of function template specialization 'convert_unary<__half, float>' requested here 588 | convert_unary<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:682:20: note: in instantiation of function template specialization 'convert_unary_cuda<__half, float>' requested here 682 | return convert_unary_cuda; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:4:33: warning: unused parameter 'p0' [-Wunused-parameter] 4 | const int s0, const int p0, const int d0, const int output_size, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:4:47: warning: unused parameter 'd0' [-Wunused-parameter] 4 | const int s0, const int p0, const int d0, const int output_size, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:5:79: warning: unused parameter 'src0_ne3' [-Wunused-parameter] 5 | const int src0_ne0, const int src0_ne1, const int src0_ne2, const int src0_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:39: warning: unused parameter 'src1_ne1' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:59: warning: unused parameter 'src1_ne2' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:6:79: warning: unused parameter 'src1_ne3' [-Wunused-parameter] 6 | const int src1_ne0, const int src1_ne1, const int src1_ne2, const int src1_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:38: warning: unused parameter 'dst_ne1' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:57: warning: unused parameter 'dst_ne2' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:7:76: warning: unused parameter 'dst_ne3' [-Wunused-parameter] 7 | const int dst_ne0, const int dst_ne1, const int dst_ne2, const int dst_ne3, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:78:19: warning: unused variable 'kernel_size' [-Wunused-variable] 78 | const int64_t kernel_size = ggml_nelements(src0); | ^~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/conv-transpose-1d.cu:79:19: warning: unused variable 'input_size' [-Wunused-variable] 79 | const int64_t input_size = ggml_nelements(src1); | ^~~~~~~~~~ 17 warnings generated when compiling for host. [ 23%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/count-equal.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/count-equal.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/count-equal.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/count-equal.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/count-equal.cu 12 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/count-equal.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/count-equal.cu:62:13: warning: 'break' will never be executed [-Wunreachable-code-break] 62 | break; | ^~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 7 warnings generated when compiling for gfx1010. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *__restrict' to 'type-parameter-0-0 *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:467:9: note: in instantiation of function template specialization 'dequantize_block_q8_0_f16' requested here 467 | dequantize_block_q8_0_f16<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:470:9: note: in instantiation of function template specialization 'dequantize_block_q8_0_f16' requested here 470 | dequantize_block_q8_0_f16<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *' to 'float *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:588:5: note: in instantiation of function template specialization 'convert_unary' requested here 588 | convert_unary<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:635:20: note: in instantiation of function template specialization 'convert_unary_cuda' requested here 635 | return convert_unary_cuda; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *' to '__half *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:588:5: note: in instantiation of function template specialization 'convert_unary<__half, float>' requested here 588 | convert_unary<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:682:20: note: in instantiation of function template specialization 'convert_unary_cuda<__half, float>' requested here 682 | return convert_unary_cuda; | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/count-equal.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 12 warnings generated when compiling for gfx1012. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/count-equal.cu:62:13: warning: 'break' will never be executed [-Wunreachable-code-break] 62 | break; | ^~~~~ 7 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/count-equal.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *__restrict' to 'type-parameter-0-0 *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:467:9: note: in instantiation of function template specialization 'dequantize_block_q8_0_f16' requested here 467 | dequantize_block_q8_0_f16<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:470:9: note: in instantiation of function template specialization 'dequantize_block_q8_0_f16' requested here 470 | dequantize_block_q8_0_f16<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *' to 'float *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:588:5: note: in instantiation of function template specialization 'convert_unary' requested here 588 | convert_unary<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:635:20: note: in instantiation of function template specialization 'convert_unary_cuda' requested here 635 | return convert_unary_cuda; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *' to '__half *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:588:5: note: in instantiation of function template specialization 'convert_unary<__half, float>' requested here 588 | convert_unary<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:682:20: note: in instantiation of function template specialization 'convert_unary_cuda<__half, float>' requested here 682 | return convert_unary_cuda; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/count-equal.cu:62:13: warning: 'break' will never be executed [-Wunreachable-code-break] 62 | break; | ^~~~~ 7 warnings generated when compiling for gfx1030. 12 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/count-equal.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/count-equal.cu:62:13: warning: 'break' will never be executed [-Wunreachable-code-break] 62 | break; | ^~~~~ 7 warnings generated when compiling for gfx1031. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *__restrict' to 'type-parameter-0-0 *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:467:9: note: in instantiation of function template specialization 'dequantize_block_q8_0_f16' requested here 467 | dequantize_block_q8_0_f16<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:470:9: note: in instantiation of function template specialization 'dequantize_block_q8_0_f16' requested here 470 | dequantize_block_q8_0_f16<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *' to 'float *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:588:5: note: in instantiation of function template specialization 'convert_unary' requested here 588 | convert_unary<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:635:20: note: in instantiation of function template specialization 'convert_unary_cuda' requested here 635 | return convert_unary_cuda; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *' to '__half *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:588:5: note: in instantiation of function template specialization 'convert_unary<__half, float>' requested here 588 | convert_unary<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:682:20: note: in instantiation of function template specialization 'convert_unary_cuda<__half, float>' requested here 682 | return convert_unary_cuda; | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/count-equal.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 12 warnings generated when compiling for gfx1031. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/count-equal.cu:62:13: warning: 'break' will never be executed [-Wunreachable-code-break] 62 | break; | ^~~~~ 7 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/count-equal.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *__restrict' to 'type-parameter-0-0 *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:467:9: note: in instantiation of function template specialization 'dequantize_block_q8_0_f16' requested here 467 | dequantize_block_q8_0_f16<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:470:9: note: in instantiation of function template specialization 'dequantize_block_q8_0_f16' requested here 470 | dequantize_block_q8_0_f16<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *' to 'float *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:588:5: note: in instantiation of function template specialization 'convert_unary' requested here 588 | convert_unary<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:635:20: note: in instantiation of function template specialization 'convert_unary_cuda' requested here 635 | return convert_unary_cuda; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *' to '__half *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:588:5: note: in instantiation of function template specialization 'convert_unary<__half, float>' requested here 588 | convert_unary<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:682:20: note: in instantiation of function template specialization 'convert_unary_cuda<__half, float>' requested here 682 | return convert_unary_cuda; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/count-equal.cu:62:13: warning: 'break' will never be executed [-Wunreachable-code-break] 62 | break; | ^~~~~ 7 warnings generated when compiling for gfx1100. 12 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/count-equal.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/count-equal.cu:62:13: warning: 'break' will never be executed [-Wunreachable-code-break] 62 | break; | ^~~~~ 7 warnings generated when compiling for gfx1101. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *__restrict' to 'type-parameter-0-0 *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:467:9: note: in instantiation of function template specialization 'dequantize_block_q8_0_f16' requested here 467 | dequantize_block_q8_0_f16<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:470:9: note: in instantiation of function template specialization 'dequantize_block_q8_0_f16' requested here 470 | dequantize_block_q8_0_f16<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *' to 'float *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:588:5: note: in instantiation of function template specialization 'convert_unary' requested here 588 | convert_unary<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:635:20: note: in instantiation of function template specialization 'convert_unary_cuda' requested here 635 | return convert_unary_cuda; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *' to '__half *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:588:5: note: in instantiation of function template specialization 'convert_unary<__half, float>' requested here 588 | convert_unary<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:682:20: note: in instantiation of function template specialization 'convert_unary_cuda<__half, float>' requested here 682 | return convert_unary_cuda; | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/count-equal.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 12 warnings generated when compiling for gfx1100. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/count-equal.cu:62:13: warning: 'break' will never be executed [-Wunreachable-code-break] 62 | break; | ^~~~~ 7 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/count-equal.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *__restrict' to 'type-parameter-0-0 *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:467:9: note: in instantiation of function template specialization 'dequantize_block_q8_0_f16' requested here 467 | dequantize_block_q8_0_f16<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:470:9: note: in instantiation of function template specialization 'dequantize_block_q8_0_f16' requested here 470 | dequantize_block_q8_0_f16<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *' to 'float *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:588:5: note: in instantiation of function template specialization 'convert_unary' requested here 588 | convert_unary<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:635:20: note: in instantiation of function template specialization 'convert_unary_cuda' requested here 635 | return convert_unary_cuda; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *' to '__half *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:588:5: note: in instantiation of function template specialization 'convert_unary<__half, float>' requested here 588 | convert_unary<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:682:20: note: in instantiation of function template specialization 'convert_unary_cuda<__half, float>' requested here 682 | return convert_unary_cuda; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/count-equal.cu:62:13: warning: 'break' will never be executed [-Wunreachable-code-break] 62 | break; | ^~~~~ 7 warnings generated when compiling for gfx1103. 12 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/count-equal.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/count-equal.cu:62:13: warning: 'break' will never be executed [-Wunreachable-code-break] 62 | break; | ^~~~~ 7 warnings generated when compiling for gfx1151. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *__restrict' to 'type-parameter-0-0 *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:467:9: note: in instantiation of function template specialization 'dequantize_block_q8_0_f16' requested here 467 | dequantize_block_q8_0_f16<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:470:9: note: in instantiation of function template specialization 'dequantize_block_q8_0_f16' requested here 470 | dequantize_block_q8_0_f16<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *' to 'float *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:588:5: note: in instantiation of function template specialization 'convert_unary' requested here 588 | convert_unary<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:635:20: note: in instantiation of function template specialization 'convert_unary_cuda' requested here 635 | return convert_unary_cuda; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *' to '__half *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:588:5: note: in instantiation of function template specialization 'convert_unary<__half, float>' requested here 588 | convert_unary<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:682:20: note: in instantiation of function template specialization 'convert_unary_cuda<__half, float>' requested here 682 | return convert_unary_cuda; | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/count-equal.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 12 warnings generated when compiling for gfx1102. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/count-equal.cu:62:13: warning: 'break' will never be executed [-Wunreachable-code-break] 62 | break; | ^~~~~ 7 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/count-equal.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *__restrict' to 'type-parameter-0-0 *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:467:9: note: in instantiation of function template specialization 'dequantize_block_q8_0_f16' requested here 467 | dequantize_block_q8_0_f16<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:470:9: note: in instantiation of function template specialization 'dequantize_block_q8_0_f16' requested here 470 | dequantize_block_q8_0_f16<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *' to 'float *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:588:5: note: in instantiation of function template specialization 'convert_unary' requested here 588 | convert_unary<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:635:20: note: in instantiation of function template specialization 'convert_unary_cuda' requested here 635 | return convert_unary_cuda; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *' to '__half *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:588:5: note: in instantiation of function template specialization 'convert_unary<__half, float>' requested here 588 | convert_unary<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:682:20: note: in instantiation of function template specialization 'convert_unary_cuda<__half, float>' requested here 682 | return convert_unary_cuda; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/count-equal.cu:62:13: warning: 'break' will never be executed [-Wunreachable-code-break] 62 | break; | ^~~~~ 7 warnings generated when compiling for gfx906. 12 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/count-equal.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/count-equal.cu:62:13: warning: 'break' will never be executed [-Wunreachable-code-break] 62 | break; | ^~~~~ 7 warnings generated when compiling for gfx908. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *__restrict' to 'type-parameter-0-0 *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:467:9: note: in instantiation of function template specialization 'dequantize_block_q8_0_f16' requested here 467 | dequantize_block_q8_0_f16<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:470:9: note: in instantiation of function template specialization 'dequantize_block_q8_0_f16' requested here 470 | dequantize_block_q8_0_f16<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *' to 'float *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:588:5: note: in instantiation of function template specialization 'convert_unary' requested here 588 | convert_unary<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:635:20: note: in instantiation of function template specialization 'convert_unary_cuda' requested here 635 | return convert_unary_cuda; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *' to '__half *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:588:5: note: in instantiation of function template specialization 'convert_unary<__half, float>' requested here 588 | convert_unary<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:682:20: note: in instantiation of function template specialization 'convert_unary_cuda<__half, float>' requested here 682 | return convert_unary_cuda; | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/count-equal.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 12 warnings generated when compiling for gfx1151. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/count-equal.cu:62:13: warning: 'break' will never be executed [-Wunreachable-code-break] 62 | break; | ^~~~~ 7 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/count-equal.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *__restrict' to 'type-parameter-0-0 *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:467:9: note: in instantiation of function template specialization 'dequantize_block_q8_0_f16' requested here 467 | dequantize_block_q8_0_f16<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:470:9: note: in instantiation of function template specialization 'dequantize_block_q8_0_f16' requested here 470 | dequantize_block_q8_0_f16<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *' to 'float *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:588:5: note: in instantiation of function template specialization 'convert_unary' requested here 588 | convert_unary<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:635:20: note: in instantiation of function template specialization 'convert_unary_cuda' requested here 635 | return convert_unary_cuda; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *' to '__half *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:588:5: note: in instantiation of function template specialization 'convert_unary<__half, float>' requested here 588 | convert_unary<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:682:20: note: in instantiation of function template specialization 'convert_unary_cuda<__half, float>' requested here 682 | return convert_unary_cuda; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/count-equal.cu:62:13: warning: 'break' will never be executed [-Wunreachable-code-break] 62 | break; | ^~~~~ 7 warnings generated when compiling for gfx90a. 12 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/count-equal.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/count-equal.cu:62:13: warning: 'break' will never be executed [-Wunreachable-code-break] 62 | break; | ^~~~~ 7 warnings generated when compiling for gfx942. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *__restrict' to 'type-parameter-0-0 *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:467:9: note: in instantiation of function template specialization 'dequantize_block_q8_0_f16' requested here 467 | dequantize_block_q8_0_f16<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:470:9: note: in instantiation of function template specialization 'dequantize_block_q8_0_f16' requested here 470 | dequantize_block_q8_0_f16<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *' to 'float *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:588:5: note: in instantiation of function template specialization 'convert_unary' requested here 588 | convert_unary<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:635:20: note: in instantiation of function template specialization 'convert_unary_cuda' requested here 635 | return convert_unary_cuda; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *' to '__half *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:588:5: note: in instantiation of function template specialization 'convert_unary<__half, float>' requested here 588 | convert_unary<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:682:20: note: in instantiation of function template specialization 'convert_unary_cuda<__half, float>' requested here 682 | return convert_unary_cuda; | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/count-equal.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 12 warnings generated when compiling for gfx906. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/count-equal.cu:62:13: warning: 'break' will never be executed [-Wunreachable-code-break] 62 | break; | ^~~~~ 7 warnings generated when compiling for host. [ 24%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/cpy.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/cpy.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/cpy.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/cpy.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cpy.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cpy.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cpy.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *__restrict' to 'type-parameter-0-0 *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:467:9: note: in instantiation of function template specialization 'dequantize_block_q8_0_f16' requested here 467 | dequantize_block_q8_0_f16<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:470:9: note: in instantiation of function template specialization 'dequantize_block_q8_0_f16' requested here 470 | dequantize_block_q8_0_f16<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *' to 'float *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:588:5: note: in instantiation of function template specialization 'convert_unary' requested here 588 | convert_unary<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:635:20: note: in instantiation of function template specialization 'convert_unary_cuda' requested here 635 | return convert_unary_cuda; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *' to '__half *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:588:5: note: in instantiation of function template specialization 'convert_unary<__half, float>' requested here 588 | convert_unary<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:682:20: note: in instantiation of function template specialization 'convert_unary_cuda<__half, float>' requested here 682 | return convert_unary_cuda; | ^ 12 warnings generated when compiling for gfx908. 6 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cpy.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cpy.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *__restrict' to 'type-parameter-0-0 *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:467:9: note: in instantiation of function template specialization 'dequantize_block_q8_0_f16' requested here 467 | dequantize_block_q8_0_f16<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:470:9: note: in instantiation of function template specialization 'dequantize_block_q8_0_f16' requested here 470 | dequantize_block_q8_0_f16<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *' to 'float *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:588:5: note: in instantiation of function template specialization 'convert_unary' requested here 588 | convert_unary<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:635:20: note: in instantiation of function template specialization 'convert_unary_cuda' requested here 635 | return convert_unary_cuda; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *' to '__half *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:588:5: note: in instantiation of function template specialization 'convert_unary<__half, float>' requested here 588 | convert_unary<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:682:20: note: in instantiation of function template specialization 'convert_unary_cuda<__half, float>' requested here 682 | return convert_unary_cuda; | ^ 12 warnings generated when compiling for gfx90a. 6 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cpy.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cpy.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *__restrict' to 'type-parameter-0-0 *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:467:9: note: in instantiation of function template specialization 'dequantize_block_q8_0_f16' requested here 467 | dequantize_block_q8_0_f16<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:470:9: note: in instantiation of function template specialization 'dequantize_block_q8_0_f16' requested here 470 | dequantize_block_q8_0_f16<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *' to 'float *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:588:5: note: in instantiation of function template specialization 'convert_unary' requested here 588 | convert_unary<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:635:20: note: in instantiation of function template specialization 'convert_unary_cuda' requested here 635 | return convert_unary_cuda; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *' to '__half *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:588:5: note: in instantiation of function template specialization 'convert_unary<__half, float>' requested here 588 | convert_unary<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:682:20: note: in instantiation of function template specialization 'convert_unary_cuda<__half, float>' requested here 682 | return convert_unary_cuda; | ^ 6 warnings generated when compiling for gfx1030. 12 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cpy.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cpy.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *__restrict' to 'type-parameter-0-0 *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:467:9: note: in instantiation of function template specialization 'dequantize_block_q8_0_f16' requested here 467 | dequantize_block_q8_0_f16<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:470:9: note: in instantiation of function template specialization 'dequantize_block_q8_0_f16' requested here 470 | dequantize_block_q8_0_f16<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *' to 'float *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:588:5: note: in instantiation of function template specialization 'convert_unary' requested here 588 | convert_unary<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:635:20: note: in instantiation of function template specialization 'convert_unary_cuda' requested here 635 | return convert_unary_cuda; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *' to '__half *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:588:5: note: in instantiation of function template specialization 'convert_unary<__half, float>' requested here 588 | convert_unary<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:682:20: note: in instantiation of function template specialization 'convert_unary_cuda<__half, float>' requested here 682 | return convert_unary_cuda; | ^ 6 warnings generated when compiling for gfx1031. 12 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cpy.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cpy.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *__restrict' to 'type-parameter-0-0 *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:467:9: note: in instantiation of function template specialization 'dequantize_block_q8_0_f16' requested here 467 | dequantize_block_q8_0_f16<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:33:31: warning: cast from 'const void *' to 'int *' drops const qualifier [-Wcast-qual] 33 | const int * x0 = ((int *) vx) + blockIdx.x * nint; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:470:9: note: in instantiation of function template specialization 'dequantize_block_q8_0_f16' requested here 470 | dequantize_block_q8_0_f16<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *' to 'float *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:588:5: note: in instantiation of function template specialization 'convert_unary' requested here 588 | convert_unary<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:635:20: note: in instantiation of function template specialization 'convert_unary_cuda' requested here 635 | return convert_unary_cuda; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:580:33: warning: cast from 'const void *' to '__half *' drops const qualifier [-Wcast-qual] 580 | const src_t * x = (src_t *) vx; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:588:5: note: in instantiation of function template specialization 'convert_unary<__half, float>' requested here 588 | convert_unary<<>>(vx, y, k); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/convert.cu:682:20: note: in instantiation of function template specialization 'convert_unary_cuda<__half, float>' requested here 682 | return convert_unary_cuda; | ^ 12 warnings generated when compiling for host. [ 25%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/cross-entropy-loss.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/cross-entropy-loss.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/cross-entropy-loss.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/cross-entropy-loss.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cross-entropy-loss.cu 6 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cross-entropy-loss.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cpy.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cpy.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1010. 6 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cross-entropy-loss.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cpy.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cpy.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cross-entropy-loss.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1101. 6 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cpy.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cpy.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cross-entropy-loss.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1031. 6 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cross-entropy-loss.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cpy.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cpy.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1035. 6 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cross-entropy-loss.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cpy.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cpy.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cross-entropy-loss.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1151. 6 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cpy.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cpy.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cross-entropy-loss.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1102. 6 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cross-entropy-loss.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cpy.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cpy.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cross-entropy-loss.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx906. 6 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cpy.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cpy.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cross-entropy-loss.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx900. 6 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cross-entropy-loss.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cpy.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cpy.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx906. 6 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cross-entropy-loss.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cpy.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cpy.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cross-entropy-loss.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx90a. 6 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cpy.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cpy.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cross-entropy-loss.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx90a. 6 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cross-entropy-loss.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cpy.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cpy.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx942. 6 warnings generated when compiling for host. [ 26%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/diagmask.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/diagmask.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/diagmask.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/diagmask.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/diagmask.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/cross-entropy-loss.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for host. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/diagmask.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/diagmask.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ [ 27%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/dmmv.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/dmmv.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/dmmv.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/dmmv.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu 6 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/diagmask.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/diagmask.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:420:37: warning: cast from 'const __half *' to '__half2 *' drops const qualifier [-Wcast-qual] 420 | const half2 x_reg = *((half2 *) &(x[ib + iqs])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:682:13: warning: 'break' will never be executed [-Wunreachable-code-break] 682 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:524:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 524 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:533:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 533 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:542:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 542 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:551:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 551 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:560:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 560 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:611:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 611 | dequantize_mul_mat_vec | ^ 15 warnings generated when compiling for gfx1010. 6 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/diagmask.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/diagmask.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:420:37: warning: cast from 'const __half *' to '__half2 *' drops const qualifier [-Wcast-qual] 420 | const half2 x_reg = *((half2 *) &(x[ib + iqs])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:682:13: warning: 'break' will never be executed [-Wunreachable-code-break] 682 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:524:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 524 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:533:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 533 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:542:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 542 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:551:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 551 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:560:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 560 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:611:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 611 | dequantize_mul_mat_vec | ^ 6 warnings generated when compiling for gfx1030. 15 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/diagmask.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/diagmask.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1031. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:420:37: warning: cast from 'const __half *' to '__half2 *' drops const qualifier [-Wcast-qual] 420 | const half2 x_reg = *((half2 *) &(x[ib + iqs])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:682:13: warning: 'break' will never be executed [-Wunreachable-code-break] 682 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:524:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 524 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:533:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 533 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:542:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 542 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:551:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 551 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:560:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 560 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:611:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 611 | dequantize_mul_mat_vec | ^ 15 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/diagmask.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/diagmask.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:420:37: warning: cast from 'const __half *' to '__half2 *' drops const qualifier [-Wcast-qual] 420 | const half2 x_reg = *((half2 *) &(x[ib + iqs])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:682:13: warning: 'break' will never be executed [-Wunreachable-code-break] 682 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:524:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 524 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:533:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 533 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:542:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 542 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:551:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 551 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:560:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 560 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:611:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 611 | dequantize_mul_mat_vec | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/diagmask.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/diagmask.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 15 warnings generated when compiling for gfx1031. 6 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/diagmask.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/diagmask.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:420:37: warning: cast from 'const __half *' to '__half2 *' drops const qualifier [-Wcast-qual] 420 | const half2 x_reg = *((half2 *) &(x[ib + iqs])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:682:13: warning: 'break' will never be executed [-Wunreachable-code-break] 682 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:524:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 524 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:533:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 533 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:542:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 542 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:551:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 551 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:560:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 560 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:611:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 611 | dequantize_mul_mat_vec | ^ 6 warnings generated when compiling for gfx1101. 15 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/diagmask.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/diagmask.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:420:37: warning: cast from 'const __half *' to '__half2 *' drops const qualifier [-Wcast-qual] 420 | const half2 x_reg = *((half2 *) &(x[ib + iqs])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:682:13: warning: 'break' will never be executed [-Wunreachable-code-break] 682 | break; | ^~~~~ 6 warnings generated when compiling for gfx1102. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:524:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 524 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:533:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 533 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:542:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 542 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:551:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 551 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:560:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 560 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:611:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 611 | dequantize_mul_mat_vec | ^ 15 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/diagmask.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/diagmask.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1103. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:420:37: warning: cast from 'const __half *' to '__half2 *' drops const qualifier [-Wcast-qual] 420 | const half2 x_reg = *((half2 *) &(x[ib + iqs])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:682:13: warning: 'break' will never be executed [-Wunreachable-code-break] 682 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:524:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 524 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:533:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 533 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:542:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 542 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:551:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 551 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:560:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 560 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:611:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 611 | dequantize_mul_mat_vec | ^ 15 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/diagmask.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/diagmask.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:420:37: warning: cast from 'const __half *' to '__half2 *' drops const qualifier [-Wcast-qual] 420 | const half2 x_reg = *((half2 *) &(x[ib + iqs])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:682:13: warning: 'break' will never be executed [-Wunreachable-code-break] 682 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:524:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 524 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:533:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 533 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:542:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 542 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:551:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 551 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:560:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 560 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:611:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 611 | dequantize_mul_mat_vec | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/diagmask.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/diagmask.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 15 warnings generated when compiling for gfx1102. 6 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/diagmask.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/diagmask.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:420:37: warning: cast from 'const __half *' to '__half2 *' drops const qualifier [-Wcast-qual] 420 | const half2 x_reg = *((half2 *) &(x[ib + iqs])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:682:13: warning: 'break' will never be executed [-Wunreachable-code-break] 682 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:524:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 524 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:533:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 533 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:542:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 542 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:551:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 551 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:560:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 560 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:611:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 611 | dequantize_mul_mat_vec | ^ 15 warnings generated when compiling for gfx1103. 6 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/diagmask.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/diagmask.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:420:37: warning: cast from 'const __half *' to '__half2 *' drops const qualifier [-Wcast-qual] 420 | const half2 x_reg = *((half2 *) &(x[ib + iqs])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:682:13: warning: 'break' will never be executed [-Wunreachable-code-break] 682 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:524:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 524 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:533:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 533 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:542:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 542 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:551:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 551 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:560:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 560 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:611:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 611 | dequantize_mul_mat_vec | ^ 6 warnings generated when compiling for gfx908. 15 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/diagmask.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/diagmask.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx90a. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:420:37: warning: cast from 'const __half *' to '__half2 *' drops const qualifier [-Wcast-qual] 420 | const half2 x_reg = *((half2 *) &(x[ib + iqs])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:682:13: warning: 'break' will never be executed [-Wunreachable-code-break] 682 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:524:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 524 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:533:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 533 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:542:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 542 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:551:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 551 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:560:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 560 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:611:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 611 | dequantize_mul_mat_vec | ^ 15 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/diagmask.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/diagmask.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:420:37: warning: cast from 'const __half *' to '__half2 *' drops const qualifier [-Wcast-qual] 420 | const half2 x_reg = *((half2 *) &(x[ib + iqs])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:682:13: warning: 'break' will never be executed [-Wunreachable-code-break] 682 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:524:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 524 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:533:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 533 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:542:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 542 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:551:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 551 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:560:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 560 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:611:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 611 | dequantize_mul_mat_vec | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/diagmask.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/diagmask.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 15 warnings generated when compiling for gfx906. 6 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/diagmask.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/diagmask.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:420:37: warning: cast from 'const __half *' to '__half2 *' drops const qualifier [-Wcast-qual] 420 | const half2 x_reg = *((half2 *) &(x[ib + iqs])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:682:13: warning: 'break' will never be executed [-Wunreachable-code-break] 682 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:524:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 524 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:533:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 533 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:542:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 542 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:551:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 551 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:560:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 560 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:611:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 611 | dequantize_mul_mat_vec | ^ 6 warnings generated when compiling for host. [ 28%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/fattn-tile-f16.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/fattn-tile-f16.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/fattn-tile-f16.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/fattn-tile-f16.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu 15 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:420:37: warning: cast from 'const __half *' to '__half2 *' drops const qualifier [-Wcast-qual] 420 | const half2 x_reg = *((half2 *) &(x[ib + iqs])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:682:13: warning: 'break' will never be executed [-Wunreachable-code-break] 682 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:24:19: warning: unused parameter 'ne00' [-Wunused-parameter] 24 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:27:19: warning: unused parameter 'ne03' [-Wunused-parameter] 27 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:28:19: warning: unused parameter 'ne10' [-Wunused-parameter] 28 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:31:19: warning: unused parameter 'ne13' [-Wunused-parameter] 31 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:32:19: warning: unused parameter 'ne31' [-Wunused-parameter] 32 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:33:19: warning: unused parameter 'nb31' [-Wunused-parameter] 33 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:36:19: warning: unused parameter 'nb03' [-Wunused-parameter] 36 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:39:19: warning: unused parameter 'nb13' [-Wunused-parameter] 39 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:40:19: warning: unused parameter 'nb21' [-Wunused-parameter] 40 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:41:19: warning: unused parameter 'nb22' [-Wunused-parameter] 41 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:42:19: warning: unused parameter 'nb23' [-Wunused-parameter] 42 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:43:19: warning: unused parameter 'ne0' [-Wunused-parameter] 43 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:44:19: warning: unused parameter 'ne1' [-Wunused-parameter] 44 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:45:19: warning: unused parameter 'ne2' [-Wunused-parameter] 45 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:46:19: warning: unused parameter 'ne3' [-Wunused-parameter] 46 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:524:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 524 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:533:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 533 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:542:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 542 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:551:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 551 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:560:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 560 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:611:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 611 | dequantize_mul_mat_vec | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:294:13: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 294 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:323:13: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<16, 4, false>' requested here 323 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:300:13: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 300 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:323:13: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<16, 4, false>' requested here 323 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:294:13: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 294 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:348:9: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<32, 1, false>' requested here 348 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:300:13: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 300 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:348:9: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<32, 1, false>' requested here 348 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ 15 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:420:37: warning: cast from 'const __half *' to '__half2 *' drops const qualifier [-Wcast-qual] 420 | const half2 x_reg = *((half2 *) &(x[ib + iqs])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:682:13: warning: 'break' will never be executed [-Wunreachable-code-break] 682 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:524:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 524 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:533:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 533 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:542:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 542 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:551:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 551 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:560:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 560 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:611:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 611 | dequantize_mul_mat_vec | ^ 15 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:420:37: warning: cast from 'const __half *' to '__half2 *' drops const qualifier [-Wcast-qual] 420 | const half2 x_reg = *((half2 *) &(x[ib + iqs])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:682:13: warning: 'break' will never be executed [-Wunreachable-code-break] 682 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:524:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 524 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:533:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 533 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:542:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 542 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:551:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 551 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:560:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 560 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:611:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 611 | dequantize_mul_mat_vec | ^ 15 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 30 warnings generated when compiling for gfx1010. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:420:37: warning: cast from 'const __half *' to '__half2 *' drops const qualifier [-Wcast-qual] 420 | const half2 x_reg = *((half2 *) &(x[ib + iqs])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:682:13: warning: 'break' will never be executed [-Wunreachable-code-break] 682 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:524:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 524 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:533:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 533 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:542:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 542 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:551:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 551 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:560:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 560 | dequantize_mul_mat_vec | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:494:53: warning: cast from 'const float *' to 'HIP_vector_type *' drops const qualifier [-Wcast-qual] 494 | const dfloat2 y_reg = *((dfloat2 *) &(y[iybs + iqs + j/qr])); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/dmmv.cu:611:5: note: in instantiation of function template specialization 'dequantize_mul_mat_vec' requested here 611 | dequantize_mul_mat_vec | ^ 15 warnings generated when compiling for host. [ 29%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/fattn-tile-f32.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/fattn-tile-f32.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/fattn-tile-f32.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/fattn-tile-f32.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:24:19: warning: unused parameter 'ne00' [-Wunused-parameter] 24 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:27:19: warning: unused parameter 'ne03' [-Wunused-parameter] 27 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:28:19: warning: unused parameter 'ne10' [-Wunused-parameter] 28 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:31:19: warning: unused parameter 'ne13' [-Wunused-parameter] 31 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:32:19: warning: unused parameter 'ne31' [-Wunused-parameter] 32 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:33:19: warning: unused parameter 'nb31' [-Wunused-parameter] 33 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:36:19: warning: unused parameter 'nb03' [-Wunused-parameter] 36 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:39:19: warning: unused parameter 'nb13' [-Wunused-parameter] 39 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:40:19: warning: unused parameter 'nb21' [-Wunused-parameter] 40 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:41:19: warning: unused parameter 'nb22' [-Wunused-parameter] 41 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:42:19: warning: unused parameter 'nb23' [-Wunused-parameter] 42 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:43:19: warning: unused parameter 'ne0' [-Wunused-parameter] 43 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:44:19: warning: unused parameter 'ne1' [-Wunused-parameter] 44 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:45:19: warning: unused parameter 'ne2' [-Wunused-parameter] 45 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:46:19: warning: unused parameter 'ne3' [-Wunused-parameter] 46 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:294:13: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 294 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:323:13: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<16, 4, false>' requested here 323 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:300:13: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 300 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:323:13: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<16, 4, false>' requested here 323 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:294:13: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 294 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:348:9: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<32, 1, false>' requested here 348 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:300:13: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 300 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:348:9: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<32, 1, false>' requested here 348 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:24:19: warning: unused parameter 'ne00' [-Wunused-parameter] 24 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:27:19: warning: unused parameter 'ne03' [-Wunused-parameter] 27 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:28:19: warning: unused parameter 'ne10' [-Wunused-parameter] 28 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:31:19: warning: unused parameter 'ne13' [-Wunused-parameter] 31 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:32:19: warning: unused parameter 'ne31' [-Wunused-parameter] 32 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:33:19: warning: unused parameter 'nb31' [-Wunused-parameter] 33 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:36:19: warning: unused parameter 'nb03' [-Wunused-parameter] 36 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:39:19: warning: unused parameter 'nb13' [-Wunused-parameter] 39 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:40:19: warning: unused parameter 'nb21' [-Wunused-parameter] 40 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:41:19: warning: unused parameter 'nb22' [-Wunused-parameter] 41 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:42:19: warning: unused parameter 'nb23' [-Wunused-parameter] 42 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:43:19: warning: unused parameter 'ne0' [-Wunused-parameter] 43 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:44:19: warning: unused parameter 'ne1' [-Wunused-parameter] 44 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:45:19: warning: unused parameter 'ne2' [-Wunused-parameter] 45 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:46:19: warning: unused parameter 'ne3' [-Wunused-parameter] 46 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:293:13: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 293 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:319:13: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<16, 4, false>' requested here 319 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:299:13: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 299 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:319:13: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<16, 4, false>' requested here 319 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:293:13: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 293 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:344:9: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<32, 1, false>' requested here 344 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:299:13: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 299 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:344:9: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<32, 1, false>' requested here 344 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ 30 warnings generated when compiling for gfx1012. 30 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:24:19: warning: unused parameter 'ne00' [-Wunused-parameter] 24 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:27:19: warning: unused parameter 'ne03' [-Wunused-parameter] 27 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:28:19: warning: unused parameter 'ne10' [-Wunused-parameter] 28 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:31:19: warning: unused parameter 'ne13' [-Wunused-parameter] 31 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:32:19: warning: unused parameter 'ne31' [-Wunused-parameter] 32 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:33:19: warning: unused parameter 'nb31' [-Wunused-parameter] 33 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:36:19: warning: unused parameter 'nb03' [-Wunused-parameter] 36 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:39:19: warning: unused parameter 'nb13' [-Wunused-parameter] 39 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:40:19: warning: unused parameter 'nb21' [-Wunused-parameter] 40 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:41:19: warning: unused parameter 'nb22' [-Wunused-parameter] 41 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:42:19: warning: unused parameter 'nb23' [-Wunused-parameter] 42 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:43:19: warning: unused parameter 'ne0' [-Wunused-parameter] 43 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:44:19: warning: unused parameter 'ne1' [-Wunused-parameter] 44 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:45:19: warning: unused parameter 'ne2' [-Wunused-parameter] 45 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:46:19: warning: unused parameter 'ne3' [-Wunused-parameter] 46 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:294:13: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 294 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:323:13: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<16, 4, false>' requested here 323 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:300:13: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 300 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:323:13: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<16, 4, false>' requested here 323 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:294:13: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 294 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:348:9: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<32, 1, false>' requested here 348 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:300:13: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 300 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:348:9: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<32, 1, false>' requested here 348 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:24:19: warning: unused parameter 'ne00' [-Wunused-parameter] 24 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:27:19: warning: unused parameter 'ne03' [-Wunused-parameter] 27 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:28:19: warning: unused parameter 'ne10' [-Wunused-parameter] 28 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:31:19: warning: unused parameter 'ne13' [-Wunused-parameter] 31 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:32:19: warning: unused parameter 'ne31' [-Wunused-parameter] 32 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:33:19: warning: unused parameter 'nb31' [-Wunused-parameter] 33 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:36:19: warning: unused parameter 'nb03' [-Wunused-parameter] 36 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:39:19: warning: unused parameter 'nb13' [-Wunused-parameter] 39 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:40:19: warning: unused parameter 'nb21' [-Wunused-parameter] 40 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:41:19: warning: unused parameter 'nb22' [-Wunused-parameter] 41 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:42:19: warning: unused parameter 'nb23' [-Wunused-parameter] 42 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:43:19: warning: unused parameter 'ne0' [-Wunused-parameter] 43 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:44:19: warning: unused parameter 'ne1' [-Wunused-parameter] 44 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:45:19: warning: unused parameter 'ne2' [-Wunused-parameter] 45 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:46:19: warning: unused parameter 'ne3' [-Wunused-parameter] 46 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:293:13: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 293 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:319:13: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<16, 4, false>' requested here 319 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:299:13: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 299 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:319:13: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<16, 4, false>' requested here 319 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:293:13: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 293 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:344:9: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<32, 1, false>' requested here 344 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:299:13: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 299 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:344:9: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<32, 1, false>' requested here 344 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ 30 warnings generated when compiling for gfx1030. 30 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:24:19: warning: unused parameter 'ne00' [-Wunused-parameter] 24 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:27:19: warning: unused parameter 'ne03' [-Wunused-parameter] 27 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:28:19: warning: unused parameter 'ne10' [-Wunused-parameter] 28 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:31:19: warning: unused parameter 'ne13' [-Wunused-parameter] 31 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:32:19: warning: unused parameter 'ne31' [-Wunused-parameter] 32 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:33:19: warning: unused parameter 'nb31' [-Wunused-parameter] 33 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:36:19: warning: unused parameter 'nb03' [-Wunused-parameter] 36 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:39:19: warning: unused parameter 'nb13' [-Wunused-parameter] 39 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:40:19: warning: unused parameter 'nb21' [-Wunused-parameter] 40 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:41:19: warning: unused parameter 'nb22' [-Wunused-parameter] 41 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:42:19: warning: unused parameter 'nb23' [-Wunused-parameter] 42 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:43:19: warning: unused parameter 'ne0' [-Wunused-parameter] 43 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:44:19: warning: unused parameter 'ne1' [-Wunused-parameter] 44 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:45:19: warning: unused parameter 'ne2' [-Wunused-parameter] 45 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:46:19: warning: unused parameter 'ne3' [-Wunused-parameter] 46 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:294:13: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 294 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:323:13: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<16, 4, false>' requested here 323 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:300:13: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 300 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:323:13: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<16, 4, false>' requested here 323 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:294:13: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 294 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:348:9: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<32, 1, false>' requested here 348 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:300:13: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 300 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:348:9: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<32, 1, false>' requested here 348 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:24:19: warning: unused parameter 'ne00' [-Wunused-parameter] 24 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:27:19: warning: unused parameter 'ne03' [-Wunused-parameter] 27 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:28:19: warning: unused parameter 'ne10' [-Wunused-parameter] 28 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:31:19: warning: unused parameter 'ne13' [-Wunused-parameter] 31 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:32:19: warning: unused parameter 'ne31' [-Wunused-parameter] 32 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:33:19: warning: unused parameter 'nb31' [-Wunused-parameter] 33 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:36:19: warning: unused parameter 'nb03' [-Wunused-parameter] 36 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:39:19: warning: unused parameter 'nb13' [-Wunused-parameter] 39 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:40:19: warning: unused parameter 'nb21' [-Wunused-parameter] 40 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:41:19: warning: unused parameter 'nb22' [-Wunused-parameter] 41 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:42:19: warning: unused parameter 'nb23' [-Wunused-parameter] 42 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:43:19: warning: unused parameter 'ne0' [-Wunused-parameter] 43 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:44:19: warning: unused parameter 'ne1' [-Wunused-parameter] 44 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:45:19: warning: unused parameter 'ne2' [-Wunused-parameter] 45 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:46:19: warning: unused parameter 'ne3' [-Wunused-parameter] 46 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:293:13: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 293 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:319:13: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<16, 4, false>' requested here 319 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:299:13: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 299 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:319:13: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<16, 4, false>' requested here 319 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:293:13: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 293 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:344:9: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<32, 1, false>' requested here 344 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:299:13: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 299 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:344:9: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<32, 1, false>' requested here 344 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ 30 warnings generated when compiling for gfx1031. 30 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:24:19: warning: unused parameter 'ne00' [-Wunused-parameter] 24 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:27:19: warning: unused parameter 'ne03' [-Wunused-parameter] 27 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:28:19: warning: unused parameter 'ne10' [-Wunused-parameter] 28 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:31:19: warning: unused parameter 'ne13' [-Wunused-parameter] 31 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:32:19: warning: unused parameter 'ne31' [-Wunused-parameter] 32 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:33:19: warning: unused parameter 'nb31' [-Wunused-parameter] 33 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:36:19: warning: unused parameter 'nb03' [-Wunused-parameter] 36 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:39:19: warning: unused parameter 'nb13' [-Wunused-parameter] 39 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:40:19: warning: unused parameter 'nb21' [-Wunused-parameter] 40 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:41:19: warning: unused parameter 'nb22' [-Wunused-parameter] 41 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:42:19: warning: unused parameter 'nb23' [-Wunused-parameter] 42 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:43:19: warning: unused parameter 'ne0' [-Wunused-parameter] 43 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:44:19: warning: unused parameter 'ne1' [-Wunused-parameter] 44 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:45:19: warning: unused parameter 'ne2' [-Wunused-parameter] 45 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:46:19: warning: unused parameter 'ne3' [-Wunused-parameter] 46 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:294:13: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 294 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:323:13: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<16, 4, false>' requested here 323 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:300:13: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 300 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:323:13: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<16, 4, false>' requested here 323 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:294:13: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 294 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:348:9: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<32, 1, false>' requested here 348 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:24:19: warning: unused parameter 'ne00' [-Wunused-parameter] 24 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:27:19: warning: unused parameter 'ne03' [-Wunused-parameter] 27 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:28:19: warning: unused parameter 'ne10' [-Wunused-parameter] 28 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:31:19: warning: unused parameter 'ne13' [-Wunused-parameter] 31 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:32:19: warning: unused parameter 'ne31' [-Wunused-parameter] 32 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:33:19: warning: unused parameter 'nb31' [-Wunused-parameter] 33 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:36:19: warning: unused parameter 'nb03' [-Wunused-parameter] 36 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:39:19: warning: unused parameter 'nb13' [-Wunused-parameter] 39 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:40:19: warning: unused parameter 'nb21' [-Wunused-parameter] 40 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:41:19: warning: unused parameter 'nb22' [-Wunused-parameter] 41 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:42:19: warning: unused parameter 'nb23' [-Wunused-parameter] 42 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:43:19: warning: unused parameter 'ne0' [-Wunused-parameter] 43 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:44:19: warning: unused parameter 'ne1' [-Wunused-parameter] 44 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:45:19: warning: unused parameter 'ne2' [-Wunused-parameter] 45 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:46:19: warning: unused parameter 'ne3' [-Wunused-parameter] 46 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:300:13: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 300 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:348:9: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<32, 1, false>' requested here 348 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:293:13: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 293 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:319:13: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<16, 4, false>' requested here 319 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:299:13: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 299 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:319:13: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<16, 4, false>' requested here 319 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:293:13: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 293 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:344:9: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<32, 1, false>' requested here 344 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:299:13: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 299 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:344:9: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<32, 1, false>' requested here 344 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ 30 warnings generated when compiling for gfx1035. 30 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:24:19: warning: unused parameter 'ne00' [-Wunused-parameter] 24 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:27:19: warning: unused parameter 'ne03' [-Wunused-parameter] 27 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:28:19: warning: unused parameter 'ne10' [-Wunused-parameter] 28 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:31:19: warning: unused parameter 'ne13' [-Wunused-parameter] 31 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:32:19: warning: unused parameter 'ne31' [-Wunused-parameter] 32 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:33:19: warning: unused parameter 'nb31' [-Wunused-parameter] 33 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:36:19: warning: unused parameter 'nb03' [-Wunused-parameter] 36 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:39:19: warning: unused parameter 'nb13' [-Wunused-parameter] 39 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:40:19: warning: unused parameter 'nb21' [-Wunused-parameter] 40 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:41:19: warning: unused parameter 'nb22' [-Wunused-parameter] 41 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:42:19: warning: unused parameter 'nb23' [-Wunused-parameter] 42 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:43:19: warning: unused parameter 'ne0' [-Wunused-parameter] 43 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:44:19: warning: unused parameter 'ne1' [-Wunused-parameter] 44 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:45:19: warning: unused parameter 'ne2' [-Wunused-parameter] 45 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:46:19: warning: unused parameter 'ne3' [-Wunused-parameter] 46 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:294:13: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 294 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:323:13: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<16, 4, false>' requested here 323 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:300:13: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 300 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:323:13: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<16, 4, false>' requested here 323 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:24:19: warning: unused parameter 'ne00' [-Wunused-parameter] 24 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:27:19: warning: unused parameter 'ne03' [-Wunused-parameter] 27 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:28:19: warning: unused parameter 'ne10' [-Wunused-parameter] 28 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:31:19: warning: unused parameter 'ne13' [-Wunused-parameter] 31 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:32:19: warning: unused parameter 'ne31' [-Wunused-parameter] 32 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:33:19: warning: unused parameter 'nb31' [-Wunused-parameter] 33 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:36:19: warning: unused parameter 'nb03' [-Wunused-parameter] 36 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:39:19: warning: unused parameter 'nb13' [-Wunused-parameter] 39 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:40:19: warning: unused parameter 'nb21' [-Wunused-parameter] 40 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:41:19: warning: unused parameter 'nb22' [-Wunused-parameter] 41 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:42:19: warning: unused parameter 'nb23' [-Wunused-parameter] 42 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:43:19: warning: unused parameter 'ne0' [-Wunused-parameter] 43 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:44:19: warning: unused parameter 'ne1' [-Wunused-parameter] 44 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:45:19: warning: unused parameter 'ne2' [-Wunused-parameter] 45 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:46:19: warning: unused parameter 'ne3' [-Wunused-parameter] 46 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:293:13: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 293 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:319:13: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<16, 4, false>' requested here 319 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:299:13: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 299 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:319:13: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<16, 4, false>' requested here 319 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:293:13: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 293 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:344:9: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<32, 1, false>' requested here 344 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:299:13: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 299 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:344:9: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<32, 1, false>' requested here 344 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:294:13: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 294 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:348:9: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<32, 1, false>' requested here 348 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:300:13: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 300 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:348:9: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<32, 1, false>' requested here 348 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ 30 warnings generated when compiling for gfx1035. 30 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:24:19: warning: unused parameter 'ne00' [-Wunused-parameter] 24 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:27:19: warning: unused parameter 'ne03' [-Wunused-parameter] 27 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:28:19: warning: unused parameter 'ne10' [-Wunused-parameter] 28 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:31:19: warning: unused parameter 'ne13' [-Wunused-parameter] 31 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:32:19: warning: unused parameter 'ne31' [-Wunused-parameter] 32 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:33:19: warning: unused parameter 'nb31' [-Wunused-parameter] 33 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:36:19: warning: unused parameter 'nb03' [-Wunused-parameter] 36 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:39:19: warning: unused parameter 'nb13' [-Wunused-parameter] 39 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:40:19: warning: unused parameter 'nb21' [-Wunused-parameter] 40 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:41:19: warning: unused parameter 'nb22' [-Wunused-parameter] 41 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:42:19: warning: unused parameter 'nb23' [-Wunused-parameter] 42 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:43:19: warning: unused parameter 'ne0' [-Wunused-parameter] 43 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:44:19: warning: unused parameter 'ne1' [-Wunused-parameter] 44 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:45:19: warning: unused parameter 'ne2' [-Wunused-parameter] 45 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:46:19: warning: unused parameter 'ne3' [-Wunused-parameter] 46 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:293:13: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 293 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:319:13: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<16, 4, false>' requested here 319 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:24:19: warning: unused parameter 'ne00' [-Wunused-parameter] 24 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:27:19: warning: unused parameter 'ne03' [-Wunused-parameter] 27 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:28:19: warning: unused parameter 'ne10' [-Wunused-parameter] 28 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:31:19: warning: unused parameter 'ne13' [-Wunused-parameter] 31 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:32:19: warning: unused parameter 'ne31' [-Wunused-parameter] 32 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:33:19: warning: unused parameter 'nb31' [-Wunused-parameter] 33 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:36:19: warning: unused parameter 'nb03' [-Wunused-parameter] 36 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:39:19: warning: unused parameter 'nb13' [-Wunused-parameter] 39 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:40:19: warning: unused parameter 'nb21' [-Wunused-parameter] 40 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:41:19: warning: unused parameter 'nb22' [-Wunused-parameter] 41 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:42:19: warning: unused parameter 'nb23' [-Wunused-parameter] 42 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:43:19: warning: unused parameter 'ne0' [-Wunused-parameter] 43 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:44:19: warning: unused parameter 'ne1' [-Wunused-parameter] 44 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:45:19: warning: unused parameter 'ne2' [-Wunused-parameter] 45 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:46:19: warning: unused parameter 'ne3' [-Wunused-parameter] 46 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:299:13: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 299 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:319:13: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<16, 4, false>' requested here 319 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:293:13: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 293 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:344:9: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<32, 1, false>' requested here 344 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:299:13: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 299 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:344:9: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<32, 1, false>' requested here 344 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:294:13: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 294 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:323:13: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<16, 4, false>' requested here 323 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:300:13: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 300 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:323:13: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<16, 4, false>' requested here 323 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:294:13: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 294 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:348:9: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<32, 1, false>' requested here 348 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:300:13: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 300 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:348:9: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<32, 1, false>' requested here 348 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ 30 warnings generated when compiling for gfx1100. 30 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:24:19: warning: unused parameter 'ne00' [-Wunused-parameter] 24 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:27:19: warning: unused parameter 'ne03' [-Wunused-parameter] 27 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:28:19: warning: unused parameter 'ne10' [-Wunused-parameter] 28 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:31:19: warning: unused parameter 'ne13' [-Wunused-parameter] 31 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:32:19: warning: unused parameter 'ne31' [-Wunused-parameter] 32 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:33:19: warning: unused parameter 'nb31' [-Wunused-parameter] 33 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:36:19: warning: unused parameter 'nb03' [-Wunused-parameter] 36 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:39:19: warning: unused parameter 'nb13' [-Wunused-parameter] 39 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:40:19: warning: unused parameter 'nb21' [-Wunused-parameter] 40 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:41:19: warning: unused parameter 'nb22' [-Wunused-parameter] 41 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:42:19: warning: unused parameter 'nb23' [-Wunused-parameter] 42 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:43:19: warning: unused parameter 'ne0' [-Wunused-parameter] 43 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:44:19: warning: unused parameter 'ne1' [-Wunused-parameter] 44 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:45:19: warning: unused parameter 'ne2' [-Wunused-parameter] 45 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:46:19: warning: unused parameter 'ne3' [-Wunused-parameter] 46 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:293:13: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 293 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:319:13: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<16, 4, false>' requested here 319 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:299:13: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 299 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:319:13: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<16, 4, false>' requested here 319 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:293:13: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 293 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:344:9: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<32, 1, false>' requested here 344 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:299:13: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 299 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:344:9: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<32, 1, false>' requested here 344 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:24:19: warning: unused parameter 'ne00' [-Wunused-parameter] 24 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:27:19: warning: unused parameter 'ne03' [-Wunused-parameter] 27 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:28:19: warning: unused parameter 'ne10' [-Wunused-parameter] 28 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:31:19: warning: unused parameter 'ne13' [-Wunused-parameter] 31 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:32:19: warning: unused parameter 'ne31' [-Wunused-parameter] 32 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:33:19: warning: unused parameter 'nb31' [-Wunused-parameter] 33 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:36:19: warning: unused parameter 'nb03' [-Wunused-parameter] 36 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:39:19: warning: unused parameter 'nb13' [-Wunused-parameter] 39 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:40:19: warning: unused parameter 'nb21' [-Wunused-parameter] 40 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:41:19: warning: unused parameter 'nb22' [-Wunused-parameter] 41 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:42:19: warning: unused parameter 'nb23' [-Wunused-parameter] 42 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:43:19: warning: unused parameter 'ne0' [-Wunused-parameter] 43 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:44:19: warning: unused parameter 'ne1' [-Wunused-parameter] 44 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:45:19: warning: unused parameter 'ne2' [-Wunused-parameter] 45 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:46:19: warning: unused parameter 'ne3' [-Wunused-parameter] 46 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:294:13: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 294 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:323:13: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<16, 4, false>' requested here 323 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:300:13: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 300 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:323:13: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<16, 4, false>' requested here 323 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:294:13: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 294 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:348:9: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<32, 1, false>' requested here 348 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:300:13: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 300 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:348:9: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<32, 1, false>' requested here 348 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ 30 warnings generated when compiling for gfx1101. 30 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:24:19: warning: unused parameter 'ne00' [-Wunused-parameter] 24 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:27:19: warning: unused parameter 'ne03' [-Wunused-parameter] 27 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:28:19: warning: unused parameter 'ne10' [-Wunused-parameter] 28 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:31:19: warning: unused parameter 'ne13' [-Wunused-parameter] 31 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:32:19: warning: unused parameter 'ne31' [-Wunused-parameter] 32 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:33:19: warning: unused parameter 'nb31' [-Wunused-parameter] 33 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:36:19: warning: unused parameter 'nb03' [-Wunused-parameter] 36 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:39:19: warning: unused parameter 'nb13' [-Wunused-parameter] 39 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:40:19: warning: unused parameter 'nb21' [-Wunused-parameter] 40 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:41:19: warning: unused parameter 'nb22' [-Wunused-parameter] 41 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:42:19: warning: unused parameter 'nb23' [-Wunused-parameter] 42 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:43:19: warning: unused parameter 'ne0' [-Wunused-parameter] 43 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:44:19: warning: unused parameter 'ne1' [-Wunused-parameter] 44 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:45:19: warning: unused parameter 'ne2' [-Wunused-parameter] 45 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:46:19: warning: unused parameter 'ne3' [-Wunused-parameter] 46 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:293:13: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 293 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:319:13: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<16, 4, false>' requested here 319 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:299:13: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 299 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:319:13: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<16, 4, false>' requested here 319 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:293:13: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 293 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:344:9: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<32, 1, false>' requested here 344 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:299:13: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 299 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:344:9: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<32, 1, false>' requested here 344 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:24:19: warning: unused parameter 'ne00' [-Wunused-parameter] 24 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:27:19: warning: unused parameter 'ne03' [-Wunused-parameter] 27 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:28:19: warning: unused parameter 'ne10' [-Wunused-parameter] 28 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:31:19: warning: unused parameter 'ne13' [-Wunused-parameter] 31 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:32:19: warning: unused parameter 'ne31' [-Wunused-parameter] 32 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:33:19: warning: unused parameter 'nb31' [-Wunused-parameter] 33 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:36:19: warning: unused parameter 'nb03' [-Wunused-parameter] 36 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:39:19: warning: unused parameter 'nb13' [-Wunused-parameter] 39 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:40:19: warning: unused parameter 'nb21' [-Wunused-parameter] 40 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:41:19: warning: unused parameter 'nb22' [-Wunused-parameter] 41 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:42:19: warning: unused parameter 'nb23' [-Wunused-parameter] 42 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:43:19: warning: unused parameter 'ne0' [-Wunused-parameter] 43 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:44:19: warning: unused parameter 'ne1' [-Wunused-parameter] 44 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:45:19: warning: unused parameter 'ne2' [-Wunused-parameter] 45 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:46:19: warning: unused parameter 'ne3' [-Wunused-parameter] 46 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:294:13: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 294 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:323:13: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<16, 4, false>' requested here 323 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:300:13: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 300 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:323:13: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<16, 4, false>' requested here 323 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:294:13: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 294 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:348:9: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<32, 1, false>' requested here 348 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:300:13: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 300 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:348:9: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<32, 1, false>' requested here 348 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ 30 warnings generated when compiling for gfx1102. 30 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:24:19: warning: unused parameter 'ne00' [-Wunused-parameter] 24 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:27:19: warning: unused parameter 'ne03' [-Wunused-parameter] 27 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:28:19: warning: unused parameter 'ne10' [-Wunused-parameter] 28 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:31:19: warning: unused parameter 'ne13' [-Wunused-parameter] 31 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:32:19: warning: unused parameter 'ne31' [-Wunused-parameter] 32 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:33:19: warning: unused parameter 'nb31' [-Wunused-parameter] 33 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:36:19: warning: unused parameter 'nb03' [-Wunused-parameter] 36 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:39:19: warning: unused parameter 'nb13' [-Wunused-parameter] 39 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:40:19: warning: unused parameter 'nb21' [-Wunused-parameter] 40 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:41:19: warning: unused parameter 'nb22' [-Wunused-parameter] 41 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:42:19: warning: unused parameter 'nb23' [-Wunused-parameter] 42 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:43:19: warning: unused parameter 'ne0' [-Wunused-parameter] 43 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:44:19: warning: unused parameter 'ne1' [-Wunused-parameter] 44 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:45:19: warning: unused parameter 'ne2' [-Wunused-parameter] 45 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:46:19: warning: unused parameter 'ne3' [-Wunused-parameter] 46 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:293:13: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 293 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:319:13: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<16, 4, false>' requested here 319 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:299:13: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 299 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:319:13: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<16, 4, false>' requested here 319 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:293:13: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 293 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:344:9: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<32, 1, false>' requested here 344 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:299:13: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 299 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:344:9: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<32, 1, false>' requested here 344 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:24:19: warning: unused parameter 'ne00' [-Wunused-parameter] 24 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:27:19: warning: unused parameter 'ne03' [-Wunused-parameter] 27 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:28:19: warning: unused parameter 'ne10' [-Wunused-parameter] 28 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:31:19: warning: unused parameter 'ne13' [-Wunused-parameter] 31 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:32:19: warning: unused parameter 'ne31' [-Wunused-parameter] 32 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:33:19: warning: unused parameter 'nb31' [-Wunused-parameter] 33 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:36:19: warning: unused parameter 'nb03' [-Wunused-parameter] 36 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:39:19: warning: unused parameter 'nb13' [-Wunused-parameter] 39 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:40:19: warning: unused parameter 'nb21' [-Wunused-parameter] 40 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:41:19: warning: unused parameter 'nb22' [-Wunused-parameter] 41 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:42:19: warning: unused parameter 'nb23' [-Wunused-parameter] 42 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:43:19: warning: unused parameter 'ne0' [-Wunused-parameter] 43 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:44:19: warning: unused parameter 'ne1' [-Wunused-parameter] 44 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:45:19: warning: unused parameter 'ne2' [-Wunused-parameter] 45 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:46:19: warning: unused parameter 'ne3' [-Wunused-parameter] 46 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:294:13: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 294 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:323:13: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<16, 4, false>' requested here 323 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:300:13: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 300 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:323:13: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<16, 4, false>' requested here 323 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:294:13: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 294 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:348:9: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<32, 1, false>' requested here 348 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:300:13: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 300 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:348:9: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<32, 1, false>' requested here 348 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ 30 warnings generated when compiling for gfx1103. 30 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:24:19: warning: unused parameter 'ne00' [-Wunused-parameter] 24 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:27:19: warning: unused parameter 'ne03' [-Wunused-parameter] 27 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:28:19: warning: unused parameter 'ne10' [-Wunused-parameter] 28 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:31:19: warning: unused parameter 'ne13' [-Wunused-parameter] 31 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:32:19: warning: unused parameter 'ne31' [-Wunused-parameter] 32 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:33:19: warning: unused parameter 'nb31' [-Wunused-parameter] 33 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:36:19: warning: unused parameter 'nb03' [-Wunused-parameter] 36 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:39:19: warning: unused parameter 'nb13' [-Wunused-parameter] 39 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:40:19: warning: unused parameter 'nb21' [-Wunused-parameter] 40 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:41:19: warning: unused parameter 'nb22' [-Wunused-parameter] 41 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:42:19: warning: unused parameter 'nb23' [-Wunused-parameter] 42 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:43:19: warning: unused parameter 'ne0' [-Wunused-parameter] 43 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:44:19: warning: unused parameter 'ne1' [-Wunused-parameter] 44 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:45:19: warning: unused parameter 'ne2' [-Wunused-parameter] 45 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:46:19: warning: unused parameter 'ne3' [-Wunused-parameter] 46 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:293:13: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 293 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:319:13: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<16, 4, false>' requested here 319 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:299:13: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 299 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:319:13: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<16, 4, false>' requested here 319 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:293:13: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 293 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:344:9: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<32, 1, false>' requested here 344 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:299:13: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 299 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:344:9: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<32, 1, false>' requested here 344 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:24:19: warning: unused parameter 'ne00' [-Wunused-parameter] 24 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:27:19: warning: unused parameter 'ne03' [-Wunused-parameter] 27 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:28:19: warning: unused parameter 'ne10' [-Wunused-parameter] 28 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:31:19: warning: unused parameter 'ne13' [-Wunused-parameter] 31 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:32:19: warning: unused parameter 'ne31' [-Wunused-parameter] 32 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:33:19: warning: unused parameter 'nb31' [-Wunused-parameter] 33 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:36:19: warning: unused parameter 'nb03' [-Wunused-parameter] 36 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:39:19: warning: unused parameter 'nb13' [-Wunused-parameter] 39 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:40:19: warning: unused parameter 'nb21' [-Wunused-parameter] 40 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:41:19: warning: unused parameter 'nb22' [-Wunused-parameter] 41 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:42:19: warning: unused parameter 'nb23' [-Wunused-parameter] 42 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:43:19: warning: unused parameter 'ne0' [-Wunused-parameter] 43 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:44:19: warning: unused parameter 'ne1' [-Wunused-parameter] 44 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:45:19: warning: unused parameter 'ne2' [-Wunused-parameter] 45 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:46:19: warning: unused parameter 'ne3' [-Wunused-parameter] 46 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:294:13: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 294 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:323:13: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<16, 4, false>' requested here 323 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:300:13: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 300 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:323:13: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<16, 4, false>' requested here 323 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:294:13: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 294 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:348:9: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<32, 1, false>' requested here 348 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:300:13: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 300 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:348:9: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<32, 1, false>' requested here 348 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ 30 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 30 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:24:19: warning: unused parameter 'ne00' [-Wunused-parameter] 24 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:27:19: warning: unused parameter 'ne03' [-Wunused-parameter] 27 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:28:19: warning: unused parameter 'ne10' [-Wunused-parameter] 28 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:31:19: warning: unused parameter 'ne13' [-Wunused-parameter] 31 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:32:19: warning: unused parameter 'ne31' [-Wunused-parameter] 32 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:33:19: warning: unused parameter 'nb31' [-Wunused-parameter] 33 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:36:19: warning: unused parameter 'nb03' [-Wunused-parameter] 36 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:39:19: warning: unused parameter 'nb13' [-Wunused-parameter] 39 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:40:19: warning: unused parameter 'nb21' [-Wunused-parameter] 40 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:41:19: warning: unused parameter 'nb22' [-Wunused-parameter] 41 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:42:19: warning: unused parameter 'nb23' [-Wunused-parameter] 42 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:43:19: warning: unused parameter 'ne0' [-Wunused-parameter] 43 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:44:19: warning: unused parameter 'ne1' [-Wunused-parameter] 44 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:45:19: warning: unused parameter 'ne2' [-Wunused-parameter] 45 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:46:19: warning: unused parameter 'ne3' [-Wunused-parameter] 46 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:293:13: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 293 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:319:13: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<16, 4, false>' requested here 319 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:299:13: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 299 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:319:13: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<16, 4, false>' requested here 319 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:293:13: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 293 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:344:9: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<32, 1, false>' requested here 344 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:299:13: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 299 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:344:9: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<32, 1, false>' requested here 344 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:24:19: warning: unused parameter 'ne00' [-Wunused-parameter] 24 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:27:19: warning: unused parameter 'ne03' [-Wunused-parameter] 27 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:28:19: warning: unused parameter 'ne10' [-Wunused-parameter] 28 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:31:19: warning: unused parameter 'ne13' [-Wunused-parameter] 31 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:32:19: warning: unused parameter 'ne31' [-Wunused-parameter] 32 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:33:19: warning: unused parameter 'nb31' [-Wunused-parameter] 33 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:36:19: warning: unused parameter 'nb03' [-Wunused-parameter] 36 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:39:19: warning: unused parameter 'nb13' [-Wunused-parameter] 39 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:40:19: warning: unused parameter 'nb21' [-Wunused-parameter] 40 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:41:19: warning: unused parameter 'nb22' [-Wunused-parameter] 41 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:42:19: warning: unused parameter 'nb23' [-Wunused-parameter] 42 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:43:19: warning: unused parameter 'ne0' [-Wunused-parameter] 43 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:44:19: warning: unused parameter 'ne1' [-Wunused-parameter] 44 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:45:19: warning: unused parameter 'ne2' [-Wunused-parameter] 45 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:46:19: warning: unused parameter 'ne3' [-Wunused-parameter] 46 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:294:13: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 294 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:323:13: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<16, 4, false>' requested here 323 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:300:13: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 300 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:323:13: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<16, 4, false>' requested here 323 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:294:13: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 294 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:348:9: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<32, 1, false>' requested here 348 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:300:13: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 300 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:348:9: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<32, 1, false>' requested here 348 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ 30 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:24:19: warning: unused parameter 'ne00' [-Wunused-parameter] 24 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:27:19: warning: unused parameter 'ne03' [-Wunused-parameter] 27 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:28:19: warning: unused parameter 'ne10' [-Wunused-parameter] 28 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:31:19: warning: unused parameter 'ne13' [-Wunused-parameter] 31 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:32:19: warning: unused parameter 'ne31' [-Wunused-parameter] 32 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:33:19: warning: unused parameter 'nb31' [-Wunused-parameter] 33 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:36:19: warning: unused parameter 'nb03' [-Wunused-parameter] 36 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:39:19: warning: unused parameter 'nb13' [-Wunused-parameter] 39 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:40:19: warning: unused parameter 'nb21' [-Wunused-parameter] 40 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:41:19: warning: unused parameter 'nb22' [-Wunused-parameter] 41 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:42:19: warning: unused parameter 'nb23' [-Wunused-parameter] 42 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:43:19: warning: unused parameter 'ne0' [-Wunused-parameter] 43 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:44:19: warning: unused parameter 'ne1' [-Wunused-parameter] 44 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:45:19: warning: unused parameter 'ne2' [-Wunused-parameter] 45 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:46:19: warning: unused parameter 'ne3' [-Wunused-parameter] 46 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:293:13: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 293 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:319:13: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<16, 4, false>' requested here 319 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:299:13: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 299 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:319:13: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<16, 4, false>' requested here 319 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:293:13: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 293 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:344:9: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<32, 1, false>' requested here 344 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:299:13: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 299 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:344:9: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<32, 1, false>' requested here 344 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ 30 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:24:19: warning: unused parameter 'ne00' [-Wunused-parameter] 24 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:27:19: warning: unused parameter 'ne03' [-Wunused-parameter] 27 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:28:19: warning: unused parameter 'ne10' [-Wunused-parameter] 28 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:31:19: warning: unused parameter 'ne13' [-Wunused-parameter] 31 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:32:19: warning: unused parameter 'ne31' [-Wunused-parameter] 32 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:33:19: warning: unused parameter 'nb31' [-Wunused-parameter] 33 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:36:19: warning: unused parameter 'nb03' [-Wunused-parameter] 36 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:39:19: warning: unused parameter 'nb13' [-Wunused-parameter] 39 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:40:19: warning: unused parameter 'nb21' [-Wunused-parameter] 40 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:41:19: warning: unused parameter 'nb22' [-Wunused-parameter] 41 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:42:19: warning: unused parameter 'nb23' [-Wunused-parameter] 42 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:43:19: warning: unused parameter 'ne0' [-Wunused-parameter] 43 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:44:19: warning: unused parameter 'ne1' [-Wunused-parameter] 44 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:45:19: warning: unused parameter 'ne2' [-Wunused-parameter] 45 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:46:19: warning: unused parameter 'ne3' [-Wunused-parameter] 46 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:294:13: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 294 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:323:13: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<16, 4, false>' requested here 323 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:300:13: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 300 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:323:13: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<16, 4, false>' requested here 323 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:294:13: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 294 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:348:9: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<32, 1, false>' requested here 348 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:300:13: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 300 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:348:9: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<32, 1, false>' requested here 348 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ 30 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:24:19: warning: unused parameter 'ne00' [-Wunused-parameter] 24 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:27:19: warning: unused parameter 'ne03' [-Wunused-parameter] 27 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:28:19: warning: unused parameter 'ne10' [-Wunused-parameter] 28 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:31:19: warning: unused parameter 'ne13' [-Wunused-parameter] 31 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:32:19: warning: unused parameter 'ne31' [-Wunused-parameter] 32 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:33:19: warning: unused parameter 'nb31' [-Wunused-parameter] 33 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:36:19: warning: unused parameter 'nb03' [-Wunused-parameter] 36 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:39:19: warning: unused parameter 'nb13' [-Wunused-parameter] 39 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:40:19: warning: unused parameter 'nb21' [-Wunused-parameter] 40 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:41:19: warning: unused parameter 'nb22' [-Wunused-parameter] 41 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:42:19: warning: unused parameter 'nb23' [-Wunused-parameter] 42 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:43:19: warning: unused parameter 'ne0' [-Wunused-parameter] 43 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:44:19: warning: unused parameter 'ne1' [-Wunused-parameter] 44 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:45:19: warning: unused parameter 'ne2' [-Wunused-parameter] 45 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:46:19: warning: unused parameter 'ne3' [-Wunused-parameter] 46 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:293:13: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 293 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:319:13: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<16, 4, false>' requested here 319 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:299:13: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 299 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:319:13: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<16, 4, false>' requested here 319 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:293:13: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 293 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:344:9: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<32, 1, false>' requested here 344 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:299:13: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 299 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:344:9: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<32, 1, false>' requested here 344 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ 30 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:24:19: warning: unused parameter 'ne00' [-Wunused-parameter] 24 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:27:19: warning: unused parameter 'ne03' [-Wunused-parameter] 27 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:28:19: warning: unused parameter 'ne10' [-Wunused-parameter] 28 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:31:19: warning: unused parameter 'ne13' [-Wunused-parameter] 31 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:32:19: warning: unused parameter 'ne31' [-Wunused-parameter] 32 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:33:19: warning: unused parameter 'nb31' [-Wunused-parameter] 33 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:36:19: warning: unused parameter 'nb03' [-Wunused-parameter] 36 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:39:19: warning: unused parameter 'nb13' [-Wunused-parameter] 39 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:40:19: warning: unused parameter 'nb21' [-Wunused-parameter] 40 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:41:19: warning: unused parameter 'nb22' [-Wunused-parameter] 41 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:42:19: warning: unused parameter 'nb23' [-Wunused-parameter] 42 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:43:19: warning: unused parameter 'ne0' [-Wunused-parameter] 43 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:44:19: warning: unused parameter 'ne1' [-Wunused-parameter] 44 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:45:19: warning: unused parameter 'ne2' [-Wunused-parameter] 45 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:46:19: warning: unused parameter 'ne3' [-Wunused-parameter] 46 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:294:13: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 294 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:323:13: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<16, 4, false>' requested here 323 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:300:13: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 300 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:323:13: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<16, 4, false>' requested here 323 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:294:13: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 294 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:348:9: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<32, 1, false>' requested here 348 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:300:13: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 300 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:348:9: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<32, 1, false>' requested here 348 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ 30 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:24:19: warning: unused parameter 'ne00' [-Wunused-parameter] 24 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:27:19: warning: unused parameter 'ne03' [-Wunused-parameter] 27 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:28:19: warning: unused parameter 'ne10' [-Wunused-parameter] 28 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:31:19: warning: unused parameter 'ne13' [-Wunused-parameter] 31 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:32:19: warning: unused parameter 'ne31' [-Wunused-parameter] 32 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:33:19: warning: unused parameter 'nb31' [-Wunused-parameter] 33 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:36:19: warning: unused parameter 'nb03' [-Wunused-parameter] 36 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:39:19: warning: unused parameter 'nb13' [-Wunused-parameter] 39 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:40:19: warning: unused parameter 'nb21' [-Wunused-parameter] 40 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:41:19: warning: unused parameter 'nb22' [-Wunused-parameter] 41 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:42:19: warning: unused parameter 'nb23' [-Wunused-parameter] 42 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:43:19: warning: unused parameter 'ne0' [-Wunused-parameter] 43 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:44:19: warning: unused parameter 'ne1' [-Wunused-parameter] 44 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:45:19: warning: unused parameter 'ne2' [-Wunused-parameter] 45 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:46:19: warning: unused parameter 'ne3' [-Wunused-parameter] 46 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:293:13: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 293 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:319:13: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<16, 4, false>' requested here 319 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:299:13: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 299 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:319:13: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<16, 4, false>' requested here 319 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:293:13: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 293 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:344:9: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<32, 1, false>' requested here 344 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:299:13: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 299 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:344:9: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<32, 1, false>' requested here 344 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ 30 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:24:19: warning: unused parameter 'ne00' [-Wunused-parameter] 24 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:27:19: warning: unused parameter 'ne03' [-Wunused-parameter] 27 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:28:19: warning: unused parameter 'ne10' [-Wunused-parameter] 28 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:31:19: warning: unused parameter 'ne13' [-Wunused-parameter] 31 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:32:19: warning: unused parameter 'ne31' [-Wunused-parameter] 32 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:33:19: warning: unused parameter 'nb31' [-Wunused-parameter] 33 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:36:19: warning: unused parameter 'nb03' [-Wunused-parameter] 36 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:39:19: warning: unused parameter 'nb13' [-Wunused-parameter] 39 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:40:19: warning: unused parameter 'nb21' [-Wunused-parameter] 40 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:41:19: warning: unused parameter 'nb22' [-Wunused-parameter] 41 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:42:19: warning: unused parameter 'nb23' [-Wunused-parameter] 42 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:43:19: warning: unused parameter 'ne0' [-Wunused-parameter] 43 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:44:19: warning: unused parameter 'ne1' [-Wunused-parameter] 44 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:45:19: warning: unused parameter 'ne2' [-Wunused-parameter] 45 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:46:19: warning: unused parameter 'ne3' [-Wunused-parameter] 46 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:294:13: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 294 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:323:13: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<16, 4, false>' requested here 323 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:300:13: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 300 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:323:13: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<16, 4, false>' requested here 323 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:294:13: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 294 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:348:9: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<32, 1, false>' requested here 348 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:300:13: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 300 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:348:9: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<32, 1, false>' requested here 348 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ 30 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:24:19: warning: unused parameter 'ne00' [-Wunused-parameter] 24 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:27:19: warning: unused parameter 'ne03' [-Wunused-parameter] 27 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:28:19: warning: unused parameter 'ne10' [-Wunused-parameter] 28 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:31:19: warning: unused parameter 'ne13' [-Wunused-parameter] 31 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:32:19: warning: unused parameter 'ne31' [-Wunused-parameter] 32 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:33:19: warning: unused parameter 'nb31' [-Wunused-parameter] 33 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:36:19: warning: unused parameter 'nb03' [-Wunused-parameter] 36 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:39:19: warning: unused parameter 'nb13' [-Wunused-parameter] 39 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:40:19: warning: unused parameter 'nb21' [-Wunused-parameter] 40 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:41:19: warning: unused parameter 'nb22' [-Wunused-parameter] 41 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:42:19: warning: unused parameter 'nb23' [-Wunused-parameter] 42 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:43:19: warning: unused parameter 'ne0' [-Wunused-parameter] 43 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:44:19: warning: unused parameter 'ne1' [-Wunused-parameter] 44 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:45:19: warning: unused parameter 'ne2' [-Wunused-parameter] 45 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:46:19: warning: unused parameter 'ne3' [-Wunused-parameter] 46 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:293:13: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 293 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:319:13: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<16, 4, false>' requested here 319 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:299:13: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 299 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:319:13: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<16, 4, false>' requested here 319 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:293:13: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 293 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:344:9: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<32, 1, false>' requested here 344 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:299:13: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 299 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:344:9: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<32, 1, false>' requested here 344 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ 30 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:24:19: warning: unused parameter 'ne00' [-Wunused-parameter] 24 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:27:19: warning: unused parameter 'ne03' [-Wunused-parameter] 27 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:28:19: warning: unused parameter 'ne10' [-Wunused-parameter] 28 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:31:19: warning: unused parameter 'ne13' [-Wunused-parameter] 31 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:32:19: warning: unused parameter 'ne31' [-Wunused-parameter] 32 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:33:19: warning: unused parameter 'nb31' [-Wunused-parameter] 33 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:36:19: warning: unused parameter 'nb03' [-Wunused-parameter] 36 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:39:19: warning: unused parameter 'nb13' [-Wunused-parameter] 39 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:40:19: warning: unused parameter 'nb21' [-Wunused-parameter] 40 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:41:19: warning: unused parameter 'nb22' [-Wunused-parameter] 41 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:42:19: warning: unused parameter 'nb23' [-Wunused-parameter] 42 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:43:19: warning: unused parameter 'ne0' [-Wunused-parameter] 43 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:44:19: warning: unused parameter 'ne1' [-Wunused-parameter] 44 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:45:19: warning: unused parameter 'ne2' [-Wunused-parameter] 45 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:46:19: warning: unused parameter 'ne3' [-Wunused-parameter] 46 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:294:13: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 294 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:323:13: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<16, 4, false>' requested here 323 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:300:13: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 300 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:323:13: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<16, 4, false>' requested here 323 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:294:13: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 294 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:348:9: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<32, 1, false>' requested here 348 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:300:13: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 300 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:348:9: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<32, 1, false>' requested here 348 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ 30 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:24:19: warning: unused parameter 'ne00' [-Wunused-parameter] 24 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:27:19: warning: unused parameter 'ne03' [-Wunused-parameter] 27 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:28:19: warning: unused parameter 'ne10' [-Wunused-parameter] 28 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:31:19: warning: unused parameter 'ne13' [-Wunused-parameter] 31 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:32:19: warning: unused parameter 'ne31' [-Wunused-parameter] 32 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:33:19: warning: unused parameter 'nb31' [-Wunused-parameter] 33 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:36:19: warning: unused parameter 'nb03' [-Wunused-parameter] 36 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:39:19: warning: unused parameter 'nb13' [-Wunused-parameter] 39 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:40:19: warning: unused parameter 'nb21' [-Wunused-parameter] 40 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:41:19: warning: unused parameter 'nb22' [-Wunused-parameter] 41 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:42:19: warning: unused parameter 'nb23' [-Wunused-parameter] 42 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:43:19: warning: unused parameter 'ne0' [-Wunused-parameter] 43 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:44:19: warning: unused parameter 'ne1' [-Wunused-parameter] 44 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:45:19: warning: unused parameter 'ne2' [-Wunused-parameter] 45 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:46:19: warning: unused parameter 'ne3' [-Wunused-parameter] 46 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:293:13: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 293 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:319:13: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<16, 4, false>' requested here 319 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:299:13: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 299 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:319:13: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<16, 4, false>' requested here 319 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:293:13: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 293 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:344:9: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<32, 1, false>' requested here 344 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:299:13: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 299 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:344:9: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<32, 1, false>' requested here 344 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ 30 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:24:19: warning: unused parameter 'ne00' [-Wunused-parameter] 24 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:27:19: warning: unused parameter 'ne03' [-Wunused-parameter] 27 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:28:19: warning: unused parameter 'ne10' [-Wunused-parameter] 28 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:31:19: warning: unused parameter 'ne13' [-Wunused-parameter] 31 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:32:19: warning: unused parameter 'ne31' [-Wunused-parameter] 32 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:33:19: warning: unused parameter 'nb31' [-Wunused-parameter] 33 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:36:19: warning: unused parameter 'nb03' [-Wunused-parameter] 36 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:39:19: warning: unused parameter 'nb13' [-Wunused-parameter] 39 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:40:19: warning: unused parameter 'nb21' [-Wunused-parameter] 40 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:41:19: warning: unused parameter 'nb22' [-Wunused-parameter] 41 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:42:19: warning: unused parameter 'nb23' [-Wunused-parameter] 42 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:43:19: warning: unused parameter 'ne0' [-Wunused-parameter] 43 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:44:19: warning: unused parameter 'ne1' [-Wunused-parameter] 44 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:45:19: warning: unused parameter 'ne2' [-Wunused-parameter] 45 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:46:19: warning: unused parameter 'ne3' [-Wunused-parameter] 46 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:294:13: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 294 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:323:13: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<16, 4, false>' requested here 323 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:300:13: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 300 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:323:13: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<16, 4, false>' requested here 323 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:294:13: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 294 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:348:9: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<32, 1, false>' requested here 348 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:300:13: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 300 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f16.cu:348:9: note: in instantiation of function template specialization 'launch_fattn_tile_f16_64_128<32, 1, false>' requested here 348 | launch_fattn_tile_f16_64_128(ctx, dst); | ^ 30 warnings generated when compiling for host. [ 30%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/fattn.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/fattn.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/fattn.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/fattn.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu 30 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:6: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:7: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:145:13: warning: 'break' will never be executed [-Wunreachable-code-break] 145 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:118:17: warning: 'break' will never be executed [-Wunreachable-code-break] 118 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:90:17: warning: 'break' will never be executed [-Wunreachable-code-break] 90 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:67:21: warning: 'break' will never be executed [-Wunreachable-code-break] 67 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:42:21: warning: 'break' will never be executed [-Wunreachable-code-break] 42 | break; | ^~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:24:19: warning: unused parameter 'ne00' [-Wunused-parameter] 24 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:27:19: warning: unused parameter 'ne03' [-Wunused-parameter] 27 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:28:19: warning: unused parameter 'ne10' [-Wunused-parameter] 28 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:31:19: warning: unused parameter 'ne13' [-Wunused-parameter] 31 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:32:19: warning: unused parameter 'ne31' [-Wunused-parameter] 32 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:33:19: warning: unused parameter 'nb31' [-Wunused-parameter] 33 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:36:19: warning: unused parameter 'nb03' [-Wunused-parameter] 36 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:39:19: warning: unused parameter 'nb13' [-Wunused-parameter] 39 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:40:19: warning: unused parameter 'nb21' [-Wunused-parameter] 40 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:41:19: warning: unused parameter 'nb22' [-Wunused-parameter] 41 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:42:19: warning: unused parameter 'nb23' [-Wunused-parameter] 42 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:43:19: warning: unused parameter 'ne0' [-Wunused-parameter] 43 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:44:19: warning: unused parameter 'ne1' [-Wunused-parameter] 44 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:45:19: warning: unused parameter 'ne2' [-Wunused-parameter] 45 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:46:19: warning: unused parameter 'ne3' [-Wunused-parameter] 46 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:293:13: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 293 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:319:13: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<16, 4, false>' requested here 319 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:299:13: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 299 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:319:13: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<16, 4, false>' requested here 319 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:293:13: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 293 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:344:9: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<32, 1, false>' requested here 344 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:299:13: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 299 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-tile-f32.cu:344:9: note: in instantiation of function template specialization 'launch_fattn_tile_f32_64_128<32, 1, false>' requested here 344 | launch_fattn_tile_f32_64_128(ctx, dst); | ^ 80 warnings generated when compiling for gfx1010. 30 warnings generated when compiling for host. [ 31%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/getrows.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/getrows.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/getrows.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/getrows.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:6: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:7: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:145:13: warning: 'break' will never be executed [-Wunreachable-code-break] 145 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:118:17: warning: 'break' will never be executed [-Wunreachable-code-break] 118 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:90:17: warning: 'break' will never be executed [-Wunreachable-code-break] 90 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:67:21: warning: 'break' will never be executed [-Wunreachable-code-break] 67 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:42:21: warning: 'break' will never be executed [-Wunreachable-code-break] 42 | break; | ^~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cu:175:13: warning: 'break' will never be executed [-Wunreachable-code-break] 175 | break; | ^~~~~ 80 warnings generated when compiling for gfx1012. 7 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cu:175:13: warning: 'break' will never be executed [-Wunreachable-code-break] 175 | break; | ^~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:6: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:7: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:145:13: warning: 'break' will never be executed [-Wunreachable-code-break] 145 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:118:17: warning: 'break' will never be executed [-Wunreachable-code-break] 118 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:90:17: warning: 'break' will never be executed [-Wunreachable-code-break] 90 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:67:21: warning: 'break' will never be executed [-Wunreachable-code-break] 67 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:42:21: warning: 'break' will never be executed [-Wunreachable-code-break] 42 | break; | ^~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ 7 warnings generated when compiling for gfx1012. 80 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cu:175:13: warning: 'break' will never be executed [-Wunreachable-code-break] 175 | break; | ^~~~~ 7 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:6: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:7: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:145:13: warning: 'break' will never be executed [-Wunreachable-code-break] 145 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:118:17: warning: 'break' will never be executed [-Wunreachable-code-break] 118 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:90:17: warning: 'break' will never be executed [-Wunreachable-code-break] 90 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:67:21: warning: 'break' will never be executed [-Wunreachable-code-break] 67 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:42:21: warning: 'break' will never be executed [-Wunreachable-code-break] 42 | break; | ^~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ 80 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cu:175:13: warning: 'break' will never be executed [-Wunreachable-code-break] 175 | break; | ^~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 7 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:6: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:7: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:145:13: warning: 'break' will never be executed [-Wunreachable-code-break] 145 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:118:17: warning: 'break' will never be executed [-Wunreachable-code-break] 118 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:90:17: warning: 'break' will never be executed [-Wunreachable-code-break] 90 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:67:21: warning: 'break' will never be executed [-Wunreachable-code-break] 67 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:42:21: warning: 'break' will never be executed [-Wunreachable-code-break] 42 | break; | ^~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 80 warnings generated when compiling for gfx1035. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cu:175:13: warning: 'break' will never be executed [-Wunreachable-code-break] 175 | break; | ^~~~~ 7 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:6: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:7: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:145:13: warning: 'break' will never be executed [-Wunreachable-code-break] 145 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:118:17: warning: 'break' will never be executed [-Wunreachable-code-break] 118 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:90:17: warning: 'break' will never be executed [-Wunreachable-code-break] 90 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:67:21: warning: 'break' will never be executed [-Wunreachable-code-break] 67 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:42:21: warning: 'break' will never be executed [-Wunreachable-code-break] 42 | break; | ^~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 80 warnings generated when compiling for gfx1100. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cu:175:13: warning: 'break' will never be executed [-Wunreachable-code-break] 175 | break; | ^~~~~ 7 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:6: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:7: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:145:13: warning: 'break' will never be executed [-Wunreachable-code-break] 145 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:118:17: warning: 'break' will never be executed [-Wunreachable-code-break] 118 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:90:17: warning: 'break' will never be executed [-Wunreachable-code-break] 90 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:67:21: warning: 'break' will never be executed [-Wunreachable-code-break] 67 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:42:21: warning: 'break' will never be executed [-Wunreachable-code-break] 42 | break; | ^~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cu:175:13: warning: 'break' will never be executed [-Wunreachable-code-break] 175 | break; | ^~~~~ 80 warnings generated when compiling for gfx1101. 7 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:6: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:7: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:145:13: warning: 'break' will never be executed [-Wunreachable-code-break] 145 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:118:17: warning: 'break' will never be executed [-Wunreachable-code-break] 118 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:90:17: warning: 'break' will never be executed [-Wunreachable-code-break] 90 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:67:21: warning: 'break' will never be executed [-Wunreachable-code-break] 67 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:42:21: warning: 'break' will never be executed [-Wunreachable-code-break] 42 | break; | ^~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cu:175:13: warning: 'break' will never be executed [-Wunreachable-code-break] 175 | break; | ^~~~~ 7 warnings generated when compiling for gfx1102. 80 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cu:175:13: warning: 'break' will never be executed [-Wunreachable-code-break] 175 | break; | ^~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:6: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:7: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:145:13: warning: 'break' will never be executed [-Wunreachable-code-break] 145 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:118:17: warning: 'break' will never be executed [-Wunreachable-code-break] 118 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:90:17: warning: 'break' will never be executed [-Wunreachable-code-break] 90 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:67:21: warning: 'break' will never be executed [-Wunreachable-code-break] 67 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:42:21: warning: 'break' will never be executed [-Wunreachable-code-break] 42 | break; | ^~~~~ 7 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ 80 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cu:175:13: warning: 'break' will never be executed [-Wunreachable-code-break] 175 | break; | ^~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 7 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:6: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:7: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:145:13: warning: 'break' will never be executed [-Wunreachable-code-break] 145 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:118:17: warning: 'break' will never be executed [-Wunreachable-code-break] 118 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:90:17: warning: 'break' will never be executed [-Wunreachable-code-break] 90 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:67:21: warning: 'break' will never be executed [-Wunreachable-code-break] 67 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:42:21: warning: 'break' will never be executed [-Wunreachable-code-break] 42 | break; | ^~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ 80 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cu:175:13: warning: 'break' will never be executed [-Wunreachable-code-break] 175 | break; | ^~~~~ 7 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:6: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:7: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:145:13: warning: 'break' will never be executed [-Wunreachable-code-break] 145 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:118:17: warning: 'break' will never be executed [-Wunreachable-code-break] 118 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:90:17: warning: 'break' will never be executed [-Wunreachable-code-break] 90 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:67:21: warning: 'break' will never be executed [-Wunreachable-code-break] 67 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:42:21: warning: 'break' will never be executed [-Wunreachable-code-break] 42 | break; | ^~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 80 warnings generated when compiling for gfx900. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cu:175:13: warning: 'break' will never be executed [-Wunreachable-code-break] 175 | break; | ^~~~~ 7 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:6: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:7: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:145:13: warning: 'break' will never be executed [-Wunreachable-code-break] 145 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:118:17: warning: 'break' will never be executed [-Wunreachable-code-break] 118 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:90:17: warning: 'break' will never be executed [-Wunreachable-code-break] 90 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:67:21: warning: 'break' will never be executed [-Wunreachable-code-break] 67 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:42:21: warning: 'break' will never be executed [-Wunreachable-code-break] 42 | break; | ^~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cu:175:13: warning: 'break' will never be executed [-Wunreachable-code-break] 175 | break; | ^~~~~ 7 warnings generated when compiling for gfx908. 80 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cu:175:13: warning: 'break' will never be executed [-Wunreachable-code-break] 175 | break; | ^~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:6: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:7: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:145:13: warning: 'break' will never be executed [-Wunreachable-code-break] 145 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:118:17: warning: 'break' will never be executed [-Wunreachable-code-break] 118 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:90:17: warning: 'break' will never be executed [-Wunreachable-code-break] 90 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:67:21: warning: 'break' will never be executed [-Wunreachable-code-break] 67 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:42:21: warning: 'break' will never be executed [-Wunreachable-code-break] 42 | break; | ^~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ 7 warnings generated when compiling for gfx90a. 80 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cu:175:13: warning: 'break' will never be executed [-Wunreachable-code-break] 175 | break; | ^~~~~ 7 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:6: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:7: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:145:13: warning: 'break' will never be executed [-Wunreachable-code-break] 145 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:118:17: warning: 'break' will never be executed [-Wunreachable-code-break] 118 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:90:17: warning: 'break' will never be executed [-Wunreachable-code-break] 90 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:67:21: warning: 'break' will never be executed [-Wunreachable-code-break] 67 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:42:21: warning: 'break' will never be executed [-Wunreachable-code-break] 42 | break; | ^~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ 80 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cu:175:13: warning: 'break' will never be executed [-Wunreachable-code-break] 175 | break; | ^~~~~ 7 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:6: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:7: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:145:13: warning: 'break' will never be executed [-Wunreachable-code-break] 145 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:118:17: warning: 'break' will never be executed [-Wunreachable-code-break] 118 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:90:17: warning: 'break' will never be executed [-Wunreachable-code-break] 90 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:67:21: warning: 'break' will never be executed [-Wunreachable-code-break] 67 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:42:21: warning: 'break' will never be executed [-Wunreachable-code-break] 42 | break; | ^~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 80 warnings generated when compiling for gfx90a. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/getrows.cu:175:13: warning: 'break' will never be executed [-Wunreachable-code-break] 175 | break; | ^~~~~ 7 warnings generated when compiling for host. [ 32%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/ggml-cuda.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/ggml-cuda.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/ggml-cuda.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/ggml-cuda.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:5: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:6: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:7: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:145:13: warning: 'break' will never be executed [-Wunreachable-code-break] 145 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:118:17: warning: 'break' will never be executed [-Wunreachable-code-break] 118 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:90:17: warning: 'break' will never be executed [-Wunreachable-code-break] 90 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:67:21: warning: 'break' will never be executed [-Wunreachable-code-break] 67 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:42:21: warning: 'break' will never be executed [-Wunreachable-code-break] 42 | break; | ^~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:23: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:23: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ 80 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml-cuda.h:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:584:12: warning: 'backend' is deprecated: use the buffer type to find the storage location of the tensor [-Wdeprecated-declarations] 584 | struct ggml_tensor { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2019:28: note: in implicit copy constructor for 'ggml_tensor' first required here 2019 | ggml_tensor src0_row = *src0; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:587:9: note: 'backend' has been explicitly marked deprecated here 587 | GGML_DEPRECATED(enum ggml_backend_type backend, "use the buffer type to find the storage location of the tensor"); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:192:61: note: expanded from macro 'GGML_DEPRECATED' 192 | # define GGML_DEPRECATED(func, hint) func __attribute__((deprecated(hint))) | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2834:52: warning: unused parameter 'buffer' [-Wunused-parameter] 2834 | bool ggml_backend_cuda_register_host_buffer(void * buffer, size_t size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2834:67: warning: unused parameter 'size' [-Wunused-parameter] 2834 | bool ggml_backend_cuda_register_host_buffer(void * buffer, size_t size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3130:13: warning: 'break' will never be executed [-Wunreachable-code-break] 3130 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3126:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3126 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3117:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3117 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3110:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3110 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3105:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3105 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3100:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3100 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3095:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3095 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3053:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3053 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3036:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3036 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2975:13: warning: 'break' will never be executed [-Wunreachable-code-break] 2975 | break; | ^~~~~ 27 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:5: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:6: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:7: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:145:13: warning: 'break' will never be executed [-Wunreachable-code-break] 145 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:118:17: warning: 'break' will never be executed [-Wunreachable-code-break] 118 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:90:17: warning: 'break' will never be executed [-Wunreachable-code-break] 90 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:67:21: warning: 'break' will never be executed [-Wunreachable-code-break] 67 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:42:21: warning: 'break' will never be executed [-Wunreachable-code-break] 42 | break; | ^~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/fattn.cu:141:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_wmma_f16_case<256, 32, __half>' requested here 141 | ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst); | ^ 80 warnings generated when compiling for host. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:23: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ [ 33%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/im2col.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/im2col.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/im2col.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/im2col.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/im2col.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:23: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml-cuda.h:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:584:12: warning: 'backend' is deprecated: use the buffer type to find the storage location of the tensor [-Wdeprecated-declarations] 584 | struct ggml_tensor { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2019:28: note: in implicit copy constructor for 'ggml_tensor' first required here 2019 | ggml_tensor src0_row = *src0; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:587:9: note: 'backend' has been explicitly marked deprecated here 587 | GGML_DEPRECATED(enum ggml_backend_type backend, "use the buffer type to find the storage location of the tensor"); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:192:61: note: expanded from macro 'GGML_DEPRECATED' 192 | # define GGML_DEPRECATED(func, hint) func __attribute__((deprecated(hint))) | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2834:52: warning: unused parameter 'buffer' [-Wunused-parameter] 2834 | bool ggml_backend_cuda_register_host_buffer(void * buffer, size_t size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2834:67: warning: unused parameter 'size' [-Wunused-parameter] 2834 | bool ggml_backend_cuda_register_host_buffer(void * buffer, size_t size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3130:13: warning: 'break' will never be executed [-Wunreachable-code-break] 3130 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3126:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3126 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3117:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3117 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3110:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3110 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3105:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3105 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3100:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3100 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3095:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3095 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3053:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3053 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3036:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3036 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2975:13: warning: 'break' will never be executed [-Wunreachable-code-break] 2975 | break; | ^~~~~ 27 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/im2col.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/im2col.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:5: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:23: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:23: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml-cuda.h:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:584:12: warning: 'backend' is deprecated: use the buffer type to find the storage location of the tensor [-Wdeprecated-declarations] 584 | struct ggml_tensor { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2019:28: note: in implicit copy constructor for 'ggml_tensor' first required here 2019 | ggml_tensor src0_row = *src0; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:587:9: note: 'backend' has been explicitly marked deprecated here 587 | GGML_DEPRECATED(enum ggml_backend_type backend, "use the buffer type to find the storage location of the tensor"); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:192:61: note: expanded from macro 'GGML_DEPRECATED' 192 | # define GGML_DEPRECATED(func, hint) func __attribute__((deprecated(hint))) | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2834:52: warning: unused parameter 'buffer' [-Wunused-parameter] 2834 | bool ggml_backend_cuda_register_host_buffer(void * buffer, size_t size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2834:67: warning: unused parameter 'size' [-Wunused-parameter] 2834 | bool ggml_backend_cuda_register_host_buffer(void * buffer, size_t size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3130:13: warning: 'break' will never be executed [-Wunreachable-code-break] 3130 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3126:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3126 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3117:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3117 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3110:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3110 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3105:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3105 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3100:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3100 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3095:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3095 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3053:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3053 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3036:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3036 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2975:13: warning: 'break' will never be executed [-Wunreachable-code-break] 2975 | break; | ^~~~~ 27 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/im2col.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/im2col.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:5: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:23: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:23: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml-cuda.h:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:584:12: warning: 'backend' is deprecated: use the buffer type to find the storage location of the tensor [-Wdeprecated-declarations] 584 | struct ggml_tensor { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2019:28: note: in implicit copy constructor for 'ggml_tensor' first required here 2019 | ggml_tensor src0_row = *src0; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:587:9: note: 'backend' has been explicitly marked deprecated here 587 | GGML_DEPRECATED(enum ggml_backend_type backend, "use the buffer type to find the storage location of the tensor"); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:192:61: note: expanded from macro 'GGML_DEPRECATED' 192 | # define GGML_DEPRECATED(func, hint) func __attribute__((deprecated(hint))) | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2834:52: warning: unused parameter 'buffer' [-Wunused-parameter] 2834 | bool ggml_backend_cuda_register_host_buffer(void * buffer, size_t size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2834:67: warning: unused parameter 'size' [-Wunused-parameter] 2834 | bool ggml_backend_cuda_register_host_buffer(void * buffer, size_t size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3130:13: warning: 'break' will never be executed [-Wunreachable-code-break] 3130 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3126:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3126 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3117:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3117 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3110:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3110 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3105:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3105 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3100:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3100 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3095:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3095 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3053:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3053 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3036:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3036 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2975:13: warning: 'break' will never be executed [-Wunreachable-code-break] 2975 | break; | ^~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/im2col.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/im2col.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 27 warnings generated when compiling for gfx1031. 6 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:5: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/im2col.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/im2col.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:23: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:23: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml-cuda.h:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:584:12: warning: 'backend' is deprecated: use the buffer type to find the storage location of the tensor [-Wdeprecated-declarations] 584 | struct ggml_tensor { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2019:28: note: in implicit copy constructor for 'ggml_tensor' first required here 2019 | ggml_tensor src0_row = *src0; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:587:9: note: 'backend' has been explicitly marked deprecated here 587 | GGML_DEPRECATED(enum ggml_backend_type backend, "use the buffer type to find the storage location of the tensor"); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:192:61: note: expanded from macro 'GGML_DEPRECATED' 192 | # define GGML_DEPRECATED(func, hint) func __attribute__((deprecated(hint))) | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2834:52: warning: unused parameter 'buffer' [-Wunused-parameter] 2834 | bool ggml_backend_cuda_register_host_buffer(void * buffer, size_t size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2834:67: warning: unused parameter 'size' [-Wunused-parameter] 2834 | bool ggml_backend_cuda_register_host_buffer(void * buffer, size_t size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3130:13: warning: 'break' will never be executed [-Wunreachable-code-break] 3130 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3126:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3126 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3117:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3117 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3110:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3110 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3105:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3105 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3100:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3100 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3095:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3095 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3053:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3053 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3036:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3036 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2975:13: warning: 'break' will never be executed [-Wunreachable-code-break] 2975 | break; | ^~~~~ 27 warnings generated when compiling for gfx1035. 6 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:5: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/im2col.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/im2col.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:23: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:23: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml-cuda.h:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:584:12: warning: 'backend' is deprecated: use the buffer type to find the storage location of the tensor [-Wdeprecated-declarations] 584 | struct ggml_tensor { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2019:28: note: in implicit copy constructor for 'ggml_tensor' first required here 2019 | ggml_tensor src0_row = *src0; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:587:9: note: 'backend' has been explicitly marked deprecated here 587 | GGML_DEPRECATED(enum ggml_backend_type backend, "use the buffer type to find the storage location of the tensor"); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:192:61: note: expanded from macro 'GGML_DEPRECATED' 192 | # define GGML_DEPRECATED(func, hint) func __attribute__((deprecated(hint))) | ^ 6 warnings generated when compiling for gfx1035. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2834:52: warning: unused parameter 'buffer' [-Wunused-parameter] 2834 | bool ggml_backend_cuda_register_host_buffer(void * buffer, size_t size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2834:67: warning: unused parameter 'size' [-Wunused-parameter] 2834 | bool ggml_backend_cuda_register_host_buffer(void * buffer, size_t size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3130:13: warning: 'break' will never be executed [-Wunreachable-code-break] 3130 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3126:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3126 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3117:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3117 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3110:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3110 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3105:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3105 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3100:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3100 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3095:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3095 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3053:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3053 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3036:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3036 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2975:13: warning: 'break' will never be executed [-Wunreachable-code-break] 2975 | break; | ^~~~~ 27 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/im2col.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/im2col.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:5: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:23: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:23: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml-cuda.h:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:584:12: warning: 'backend' is deprecated: use the buffer type to find the storage location of the tensor [-Wdeprecated-declarations] 584 | struct ggml_tensor { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2019:28: note: in implicit copy constructor for 'ggml_tensor' first required here 2019 | ggml_tensor src0_row = *src0; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:587:9: note: 'backend' has been explicitly marked deprecated here 587 | GGML_DEPRECATED(enum ggml_backend_type backend, "use the buffer type to find the storage location of the tensor"); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:192:61: note: expanded from macro 'GGML_DEPRECATED' 192 | # define GGML_DEPRECATED(func, hint) func __attribute__((deprecated(hint))) | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2834:52: warning: unused parameter 'buffer' [-Wunused-parameter] 2834 | bool ggml_backend_cuda_register_host_buffer(void * buffer, size_t size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2834:67: warning: unused parameter 'size' [-Wunused-parameter] 2834 | bool ggml_backend_cuda_register_host_buffer(void * buffer, size_t size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3130:13: warning: 'break' will never be executed [-Wunreachable-code-break] 3130 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3126:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3126 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3117:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3117 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3110:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3110 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3105:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3105 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3100:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3100 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3095:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3095 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3053:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3053 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3036:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3036 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2975:13: warning: 'break' will never be executed [-Wunreachable-code-break] 2975 | break; | ^~~~~ 27 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/im2col.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/im2col.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:5: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:23: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:23: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml-cuda.h:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:584:12: warning: 'backend' is deprecated: use the buffer type to find the storage location of the tensor [-Wdeprecated-declarations] 584 | struct ggml_tensor { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2019:28: note: in implicit copy constructor for 'ggml_tensor' first required here 2019 | ggml_tensor src0_row = *src0; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:587:9: note: 'backend' has been explicitly marked deprecated here 587 | GGML_DEPRECATED(enum ggml_backend_type backend, "use the buffer type to find the storage location of the tensor"); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:192:61: note: expanded from macro 'GGML_DEPRECATED' 192 | # define GGML_DEPRECATED(func, hint) func __attribute__((deprecated(hint))) | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2834:52: warning: unused parameter 'buffer' [-Wunused-parameter] 2834 | bool ggml_backend_cuda_register_host_buffer(void * buffer, size_t size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2834:67: warning: unused parameter 'size' [-Wunused-parameter] 2834 | bool ggml_backend_cuda_register_host_buffer(void * buffer, size_t size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3130:13: warning: 'break' will never be executed [-Wunreachable-code-break] 3130 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3126:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3126 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3117:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3117 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3110:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3110 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3105:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3105 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3100:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3100 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3095:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3095 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3053:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3053 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3036:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3036 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2975:13: warning: 'break' will never be executed [-Wunreachable-code-break] 2975 | break; | ^~~~~ 27 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/im2col.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/im2col.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:5: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:23: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:23: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/im2col.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/im2col.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml-cuda.h:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:584:12: warning: 'backend' is deprecated: use the buffer type to find the storage location of the tensor [-Wdeprecated-declarations] 584 | struct ggml_tensor { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2019:28: note: in implicit copy constructor for 'ggml_tensor' first required here 2019 | ggml_tensor src0_row = *src0; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:587:9: note: 'backend' has been explicitly marked deprecated here 587 | GGML_DEPRECATED(enum ggml_backend_type backend, "use the buffer type to find the storage location of the tensor"); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:192:61: note: expanded from macro 'GGML_DEPRECATED' 192 | # define GGML_DEPRECATED(func, hint) func __attribute__((deprecated(hint))) | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2834:52: warning: unused parameter 'buffer' [-Wunused-parameter] 2834 | bool ggml_backend_cuda_register_host_buffer(void * buffer, size_t size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2834:67: warning: unused parameter 'size' [-Wunused-parameter] 2834 | bool ggml_backend_cuda_register_host_buffer(void * buffer, size_t size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3130:13: warning: 'break' will never be executed [-Wunreachable-code-break] 3130 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3126:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3126 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3117:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3117 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3110:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3110 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3105:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3105 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3100:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3100 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3095:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3095 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3053:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3053 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3036:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3036 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2975:13: warning: 'break' will never be executed [-Wunreachable-code-break] 2975 | break; | ^~~~~ 27 warnings generated when compiling for gfx1103. 6 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:5: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/im2col.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/im2col.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:23: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:23: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml-cuda.h:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:584:12: warning: 'backend' is deprecated: use the buffer type to find the storage location of the tensor [-Wdeprecated-declarations] 584 | struct ggml_tensor { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2019:28: note: in implicit copy constructor for 'ggml_tensor' first required here 2019 | ggml_tensor src0_row = *src0; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:587:9: note: 'backend' has been explicitly marked deprecated here 587 | GGML_DEPRECATED(enum ggml_backend_type backend, "use the buffer type to find the storage location of the tensor"); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:192:61: note: expanded from macro 'GGML_DEPRECATED' 192 | # define GGML_DEPRECATED(func, hint) func __attribute__((deprecated(hint))) | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2834:52: warning: unused parameter 'buffer' [-Wunused-parameter] 2834 | bool ggml_backend_cuda_register_host_buffer(void * buffer, size_t size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2834:67: warning: unused parameter 'size' [-Wunused-parameter] 2834 | bool ggml_backend_cuda_register_host_buffer(void * buffer, size_t size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3130:13: warning: 'break' will never be executed [-Wunreachable-code-break] 3130 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3126:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3126 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3117:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3117 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3110:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3110 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3105:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3105 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3100:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3100 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3095:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3095 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3053:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3053 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3036:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3036 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2975:13: warning: 'break' will never be executed [-Wunreachable-code-break] 2975 | break; | ^~~~~ 6 warnings generated when compiling for gfx1151. 27 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/im2col.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/im2col.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:5: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:23: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:23: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ 6 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml-cuda.h:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:584:12: warning: 'backend' is deprecated: use the buffer type to find the storage location of the tensor [-Wdeprecated-declarations] 584 | struct ggml_tensor { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2019:28: note: in implicit copy constructor for 'ggml_tensor' first required here 2019 | ggml_tensor src0_row = *src0; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:587:9: note: 'backend' has been explicitly marked deprecated here 587 | GGML_DEPRECATED(enum ggml_backend_type backend, "use the buffer type to find the storage location of the tensor"); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:192:61: note: expanded from macro 'GGML_DEPRECATED' 192 | # define GGML_DEPRECATED(func, hint) func __attribute__((deprecated(hint))) | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2834:52: warning: unused parameter 'buffer' [-Wunused-parameter] 2834 | bool ggml_backend_cuda_register_host_buffer(void * buffer, size_t size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2834:67: warning: unused parameter 'size' [-Wunused-parameter] 2834 | bool ggml_backend_cuda_register_host_buffer(void * buffer, size_t size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3130:13: warning: 'break' will never be executed [-Wunreachable-code-break] 3130 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3126:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3126 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3117:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3117 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3110:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3110 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3105:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3105 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3100:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3100 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3095:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3095 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3053:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3053 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3036:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3036 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2975:13: warning: 'break' will never be executed [-Wunreachable-code-break] 2975 | break; | ^~~~~ 27 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/im2col.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/im2col.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:5: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:23: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:23: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml-cuda.h:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:584:12: warning: 'backend' is deprecated: use the buffer type to find the storage location of the tensor [-Wdeprecated-declarations] 584 | struct ggml_tensor { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2019:28: note: in implicit copy constructor for 'ggml_tensor' first required here 2019 | ggml_tensor src0_row = *src0; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:587:9: note: 'backend' has been explicitly marked deprecated here 587 | GGML_DEPRECATED(enum ggml_backend_type backend, "use the buffer type to find the storage location of the tensor"); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:192:61: note: expanded from macro 'GGML_DEPRECATED' 192 | # define GGML_DEPRECATED(func, hint) func __attribute__((deprecated(hint))) | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2834:52: warning: unused parameter 'buffer' [-Wunused-parameter] 2834 | bool ggml_backend_cuda_register_host_buffer(void * buffer, size_t size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2834:67: warning: unused parameter 'size' [-Wunused-parameter] 2834 | bool ggml_backend_cuda_register_host_buffer(void * buffer, size_t size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3130:13: warning: 'break' will never be executed [-Wunreachable-code-break] 3130 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3126:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3126 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3117:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3117 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3110:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3110 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3105:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3105 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3100:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3100 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3095:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3095 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3053:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3053 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3036:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3036 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2975:13: warning: 'break' will never be executed [-Wunreachable-code-break] 2975 | break; | ^~~~~ 27 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/im2col.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/im2col.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:5: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:23: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:23: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml-cuda.h:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:584:12: warning: 'backend' is deprecated: use the buffer type to find the storage location of the tensor [-Wdeprecated-declarations] 584 | struct ggml_tensor { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2019:28: note: in implicit copy constructor for 'ggml_tensor' first required here 2019 | ggml_tensor src0_row = *src0; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:587:9: note: 'backend' has been explicitly marked deprecated here 587 | GGML_DEPRECATED(enum ggml_backend_type backend, "use the buffer type to find the storage location of the tensor"); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:192:61: note: expanded from macro 'GGML_DEPRECATED' 192 | # define GGML_DEPRECATED(func, hint) func __attribute__((deprecated(hint))) | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2834:52: warning: unused parameter 'buffer' [-Wunused-parameter] 2834 | bool ggml_backend_cuda_register_host_buffer(void * buffer, size_t size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2834:67: warning: unused parameter 'size' [-Wunused-parameter] 2834 | bool ggml_backend_cuda_register_host_buffer(void * buffer, size_t size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3130:13: warning: 'break' will never be executed [-Wunreachable-code-break] 3130 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3126:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3126 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3117:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3117 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3110:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3110 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3105:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3105 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3100:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3100 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3095:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3095 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3053:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3053 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3036:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3036 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2975:13: warning: 'break' will never be executed [-Wunreachable-code-break] 2975 | break; | ^~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/im2col.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/im2col.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 27 warnings generated when compiling for gfx908. 6 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:5: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:23: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:23: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/im2col.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/im2col.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml-cuda.h:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:584:12: warning: 'backend' is deprecated: use the buffer type to find the storage location of the tensor [-Wdeprecated-declarations] 584 | struct ggml_tensor { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2019:28: note: in implicit copy constructor for 'ggml_tensor' first required here 2019 | ggml_tensor src0_row = *src0; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:587:9: note: 'backend' has been explicitly marked deprecated here 587 | GGML_DEPRECATED(enum ggml_backend_type backend, "use the buffer type to find the storage location of the tensor"); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:192:61: note: expanded from macro 'GGML_DEPRECATED' 192 | # define GGML_DEPRECATED(func, hint) func __attribute__((deprecated(hint))) | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2834:52: warning: unused parameter 'buffer' [-Wunused-parameter] 2834 | bool ggml_backend_cuda_register_host_buffer(void * buffer, size_t size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2834:67: warning: unused parameter 'size' [-Wunused-parameter] 2834 | bool ggml_backend_cuda_register_host_buffer(void * buffer, size_t size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3130:13: warning: 'break' will never be executed [-Wunreachable-code-break] 3130 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3126:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3126 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3117:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3117 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3110:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3110 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3105:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3105 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3100:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3100 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3095:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3095 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3053:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3053 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3036:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3036 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2975:13: warning: 'break' will never be executed [-Wunreachable-code-break] 2975 | break; | ^~~~~ 27 warnings generated when compiling for gfx90a. 6 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:5: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/im2col.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/im2col.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:23: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:23: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml-cuda.h:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:584:12: warning: 'backend' is deprecated: use the buffer type to find the storage location of the tensor [-Wdeprecated-declarations] 584 | struct ggml_tensor { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2019:28: note: in implicit copy constructor for 'ggml_tensor' first required here 2019 | ggml_tensor src0_row = *src0; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:587:9: note: 'backend' has been explicitly marked deprecated here 587 | GGML_DEPRECATED(enum ggml_backend_type backend, "use the buffer type to find the storage location of the tensor"); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:192:61: note: expanded from macro 'GGML_DEPRECATED' 192 | # define GGML_DEPRECATED(func, hint) func __attribute__((deprecated(hint))) | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2834:52: warning: unused parameter 'buffer' [-Wunused-parameter] 2834 | bool ggml_backend_cuda_register_host_buffer(void * buffer, size_t size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2834:67: warning: unused parameter 'size' [-Wunused-parameter] 2834 | bool ggml_backend_cuda_register_host_buffer(void * buffer, size_t size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3130:13: warning: 'break' will never be executed [-Wunreachable-code-break] 3130 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3126:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3126 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3117:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3117 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3110:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3110 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3105:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3105 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3100:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3100 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3095:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3095 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3053:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3053 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3036:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3036 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2975:13: warning: 'break' will never be executed [-Wunreachable-code-break] 2975 | break; | ^~~~~ 6 warnings generated when compiling for gfx942. 27 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/im2col.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/im2col.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:5: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for host. [ 34%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/mmq.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/mmq.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/mmq.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/mmq.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:23: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:23: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml-cuda.h:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:584:12: warning: 'backend' is deprecated: use the buffer type to find the storage location of the tensor [-Wdeprecated-declarations] 584 | struct ggml_tensor { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2019:28: note: in implicit copy constructor for 'ggml_tensor' first required here 2019 | ggml_tensor src0_row = *src0; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:587:9: note: 'backend' has been explicitly marked deprecated here 587 | GGML_DEPRECATED(enum ggml_backend_type backend, "use the buffer type to find the storage location of the tensor"); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:192:61: note: expanded from macro 'GGML_DEPRECATED' 192 | # define GGML_DEPRECATED(func, hint) func __attribute__((deprecated(hint))) | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2834:52: warning: unused parameter 'buffer' [-Wunused-parameter] 2834 | bool ggml_backend_cuda_register_host_buffer(void * buffer, size_t size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2834:67: warning: unused parameter 'size' [-Wunused-parameter] 2834 | bool ggml_backend_cuda_register_host_buffer(void * buffer, size_t size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3130:13: warning: 'break' will never be executed [-Wunreachable-code-break] 3130 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3126:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3126 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3117:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3117 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3110:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3110 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3105:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3105 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3100:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3100 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3095:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3095 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3053:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3053 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3036:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3036 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2975:13: warning: 'break' will never be executed [-Wunreachable-code-break] 2975 | break; | ^~~~~ 27 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:5: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:90:13: warning: 'break' will never be executed [-Wunreachable-code-break] 90 | break; | ^~~~~ 15 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:23: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:23: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml-cuda.h:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:584:12: warning: 'backend' is deprecated: use the buffer type to find the storage location of the tensor [-Wdeprecated-declarations] 584 | struct ggml_tensor { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2019:28: note: in implicit copy constructor for 'ggml_tensor' first required here 2019 | ggml_tensor src0_row = *src0; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:587:9: note: 'backend' has been explicitly marked deprecated here 587 | GGML_DEPRECATED(enum ggml_backend_type backend, "use the buffer type to find the storage location of the tensor"); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include/ggml.h:192:61: note: expanded from macro 'GGML_DEPRECATED' 192 | # define GGML_DEPRECATED(func, hint) func __attribute__((deprecated(hint))) | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2834:52: warning: unused parameter 'buffer' [-Wunused-parameter] 2834 | bool ggml_backend_cuda_register_host_buffer(void * buffer, size_t size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2834:67: warning: unused parameter 'size' [-Wunused-parameter] 2834 | bool ggml_backend_cuda_register_host_buffer(void * buffer, size_t size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3130:13: warning: 'break' will never be executed [-Wunreachable-code-break] 3130 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3126:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3126 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3117:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3117 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3110:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3110 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3105:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3105 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3100:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3100 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3095:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3095 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3053:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3053 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:3036:15: warning: 'break' will never be executed [-Wunreachable-code-break] 3036 | } break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/ggml-cuda.cu:2975:13: warning: 'break' will never be executed [-Wunreachable-code-break] 2975 | break; | ^~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 27 warnings generated when compiling for host. [ 35%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/mmvq.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/mmvq.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/mmvq.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/mmvq.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:90:13: warning: 'break' will never be executed [-Wunreachable-code-break] 90 | break; | ^~~~~ 15 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:57:34: warning: unused parameter 'nrows_x' [-Wunused-parameter] 57 | const int ncols_x, const int nrows_x, const int nrows_y, const int nrows_dst) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:417:13: warning: 'break' will never be executed [-Wunreachable-code-break] 417 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, streIn file included from am/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu):;1 : In file included from | /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh ^: 5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_v/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ ec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ 1704:99: warning: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ unused parameter 'k00' [-Wunused-parameter] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2821/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ :15:/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ warning: unused variable 'nsm' [-Wunused-variable] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ 2821 | /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:90:13: warning: 'break' will never be executed [-Wunreachable-code-break] 90 | brea/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cuk:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ ; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ 15 warnings generated when compiling for gfx1030. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:90:13: warning: 'break' will never be executed [-Wunreachable-code-break] 90 | break; | ^~~~~ 15 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:90:13: warning: 'break' will never be executed [-Wunreachable-code-break] 90 | break; | ^~~~~ 15 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:90:13: warning: 'break' will never be executed [-Wunreachable-code-break] 90 | break; | ^~~~~ 15 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:90:13: warning: 'break' will never be executed [-Wunreachable-code-break] 90 | break; | ^~~~~ 15 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:90:13: warning: 'break' will never be executed [-Wunreachable-code-break] 90 | break; | ^~~~~ 15 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:90:13: warning: 'break' will never be executed [-Wunreachable-code-break] 90 | break; | ^~~~~ 15 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:90:13: warning: 'break' will never be executed [-Wunreachable-code-break] 90 | break; | ^~~~~ 15 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:90:13: warning: 'break' will never be executed [-Wunreachable-code-break] 90 | break; | ^~~~~ 15 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:90:13: warning: 'break' will never be executed [-Wunreachable-code-break] 90 | break; | ^~~~~ 15 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:90:13: warning: 'break' will never be executed [-Wunreachable-code-break] 90 | break; | ^~~~~ 15 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:90:13: warning: 'break' will never be executed [-Wunreachable-code-break] 90 | break; | ^~~~~ 15 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:90:13: warning: 'break' will never be executed [-Wunreachable-code-break] 90 | break; | ^~~~~ 15 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:90:13: warning: 'break' will never be executed [-Wunreachable-code-break] 90 | break; | ^~~~~ 15 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:1: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cu:90:13: warning: 'break' will never be executed [-Wunreachable-code-break] 90 | break; | ^~~~~ 15 warnings generated when compiling for host. [ 36%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/norm.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/norm.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/norm.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/norm.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/norm.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/norm.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/norm.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/norm.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/norm.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/norm.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/norm.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/norm.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/norm.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/norm.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/norm.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/norm.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/norm.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/norm.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/norm.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1101. 293 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/norm.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/norm.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:57:34: warning: unused parameter 'nrows_x' [-Wunused-parameter] 57 | const int ncols_x, const int nrows_x, const int nrows_y, const int nrows_dst) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:417:13: warning: 'break' will never be executed [-Wunreachable-code-break] 417 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/norm.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/norm.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ vec_q<<>>(vx, vy, dst, ncols_x, nrow/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.hs_:x305,: 9n:r owarning: wanonymous types declared in an anonymous union are an extension [-Wnested-anon-types]s _y, n r305o | w s _ d s t ) ; s t| r ^u ct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ 6 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/norm.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/norm.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/norm.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/norm.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/norm.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/norm.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/norm.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/norm.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/norm.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/norm.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/norm.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/norm.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/norm.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/norm.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/norm.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/norm.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for host. [ 37%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/opt-step-adamw.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/opt-step-adamw.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/opt-step-adamw.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/opt-step-adamw.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/opt-step-adamw.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/opt-step-adamw.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/opt-step-adamw.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/opt-step-adamw.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/opt-step-adamw.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/opt-step-adamw.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/opt-step-adamw.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/opt-step-adamw.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/opt-step-adamw.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/opt-step-adamw.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/opt-step-adamw.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/opt-step-adamw.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/opt-step-adamw.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/opt-step-adamw.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/opt-step-adamw.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/opt-step-adamw.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/opt-step-adamw.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/opt-step-adamw.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/opt-step-adamw.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/opt-step-adamw.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/opt-step-adamw.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/opt-step-adamw.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/opt-step-adamw.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/opt-step-adamw.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/opt-step-adamw.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/opt-step-adamw.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/opt-step-adamw.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/opt-step-adamw.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/opt-step-adamw.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/opt-step-adamw.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/opt-step-adamw.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/opt-step-adamw.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/opt-step-adamw.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/opt-step-adamw.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/opt-step-adamw.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for host. [ 38%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/out-prod.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/out-prod.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/out-prod.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/out-prod.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/out-prod.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/out-prod.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/out-prod.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/out-prod.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/out-prod.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/out-prod.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/out-prod.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/out-prod.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/out-prod.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/out-prod.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/out-prod.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 293 warnings generated when compiling for gfx1012. 6 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/out-prod.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/out-prod.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:57:34: warning: unused parameter 'nrows_x' [-Wunused-parameter] 57 | const int ncols_x, const int nrows_x, const int nrows_y, const int nrows_dst) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:417:13: warning: 'break' will never be executed [-Wunreachable-code-break] 417 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ 6 warnings generated when compiling for gfx1100. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/out-prod.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/out-prod.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/out-prod.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/out-prod.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/out-prod.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/out-prod.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/out-prod.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/out-prod.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/out-prod.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/out-prod.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/out-prod.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/out-prod.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/out-prod.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/out-prod.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx908. 161 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/out-prod.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/out-prod.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:57:34: warning: unused parameter 'nrows_x' [-Wunused-parameter] 57 | const int ncols_x, const int nrows_x, const int nrows_y, const int nrows_dst) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:417:13: warning: 'break' will never be executed [-Wunreachable-code-break] 417 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/out-prod.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/out-prod.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<< > > ( v x , vsyt,r udcstt ,{ n cols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<< > > ( vsxt,r uvcyt, {d s t| , ^ ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ 6 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/out-prod.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/out-prod.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/out-prod.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/out-prod.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for host. [ 39%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/pad.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/pad.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/pad.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/pad.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:17:56: warning: comparison of integers of different signs: 'unsigned int' and 'int' [-Wsign-compare] 17 | if (nidx < ne00 && blockIdx.y < ne01 && blockIdx.z < ne02*ne03) { | ~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:17:35: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 17 | if (nidx < ne00 && blockIdx.y < ne01 && blockIdx.z < ne02*ne03) { | ~~~~~~~~~~ ^ ~~~~ 8 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:17:56: warning: comparison of integers of different signs: 'unsigned int' and 'int' [-Wsign-compare] 17 | if (nidx < ne00 && blockIdx.y < ne01 && blockIdx.z < ne02*ne03) { | ~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:17:35: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 17 | if (nidx < ne00 && blockIdx.y < ne01 && blockIdx.z < ne02*ne03) { | ~~~~~~~~~~ ^ ~~~~ 8 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:17:56: warning: comparison of integers of different signs: 'unsigned int' and 'int' [-Wsign-compare] 17 | if (nidx < ne00 && blockIdx.y < ne01 && blockIdx.z < ne02*ne03) { | ~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:17:35: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 17 | if (nidx < ne00 && blockIdx.y < ne01 && blockIdx.z < ne02*ne03) { | ~~~~~~~~~~ ^ ~~~~ 8 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:17:56: warning: comparison of integers of different signs: 'unsigned int' and 'int' [-Wsign-compare] 17 | if (nidx < ne00 && blockIdx.y < ne01 && blockIdx.z < ne02*ne03) { | ~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:17:35: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 17 | if (nidx < ne00 && blockIdx.y < ne01 && blockIdx.z < ne02*ne03) { | ~~~~~~~~~~ ^ ~~~~ 8 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:17:56: warning: comparison of integers of different signs: 'unsigned int' and 'int' [-Wsign-compare] 17 | if (nidx < ne00 && blockIdx.y < ne01 && blockIdx.z < ne02*ne03) { | ~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:17:35: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 17 | if (nidx < ne00 && blockIdx.y < ne01 && blockIdx.z < ne02*ne03) { | ~~~~~~~~~~ ^ ~~~~ 8 warnings generated when compiling for gfx1035. 161 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:17:56: warning: comparison of integers of different signs: 'unsigned int' and 'int' [-Wsign-compare] 17 | if (nidx < ne00 && blockIdx.y < ne01 && blockIdx.z < ne02*ne03) { | ~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:17:35: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 17 | if (nidx < ne00 && blockIdx.y < ne01 && blockIdx.z < ne02*ne03) { | ~~~~~~~~~~ ^ ~~~~ 8 warnings generated when compiling for gfx1100. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:57:34: warning: unused parameter 'nrows_x' [-Wunused-parameter] 57 | const int ncols_x, const int nrows_x, const int nrows_y, const int nrows_dst) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:417:13: warning: 'break' will never be executed [-Wunreachable-code-break] 417 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:17:56: warning: comparison of integers of different signs: 'unsigned int' and 'int' [-Wsign-compare] 17 | if (nidx < ne00 && blockIdx.y < ne01 && blockIdx.z < ne02*ne03) { | ~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:17:35: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 17 | if (nidx < ne00 && blockIdx.y < ne01 && blockIdx.z < ne02*ne03) { | ~~~~~~~~~~ ^ ~~~~ 8 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:17:56: warning: comparison of integers of different signs: 'unsigned int' and 'int' [-Wsign-compare] 17 | if (nidx < ne00 && blockIdx.y < ne01 && blockIdx.z < ne02*ne03) { | ~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:17:35: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 17 | if (nidx < ne00 && blockIdx.y < ne01 && blockIdx.z < ne02*ne03) { | ~~~~~~~~~~ ^ ~~~~ 8 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:17:56: warning: comparison of integers of different signs: 'unsigned int' and 'int' [-Wsign-compare] 17 | if (nidx < ne00 && blockIdx.y < ne01 && blockIdx.z < ne02*ne03) { | ~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:17:35: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 17 | if (nidx < ne00 && blockIdx.y < ne01 && blockIdx.z < ne02*ne03) { | ~~~~~~~~~~ ^ ~~~~ 8 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:17:56: warning: comparison of integers of different signs: 'unsigned int' and 'int' [-Wsign-compare] 17 | if (nidx < ne00 && blockIdx.y < ne01 && blockIdx.z < ne02*ne03) { | ~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:17:35: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 17 | if (nidx < ne00 && blockIdx.y < ne01 && blockIdx.z < ne02*ne03) { | ~~~~~~~~~~ ^ ~~~~ 8 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:17:56: warning: comparison of integers of different signs: 'unsigned int' and 'int' [-Wsign-compare] 17 | if (nidx < ne00 && blockIdx.y < ne01 && blockIdx.z < ne02*ne03) { | ~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:17:35: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 17 | if (nidx < ne00 && blockIdx.y < ne01 && blockIdx.z < ne02*ne03) { | ~~~~~~~~~~ ^ ~~~~ 8 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:17:56: warning: comparison of integers of different signs: 'unsigned int' and 'int' [-Wsign-compare] 17 | if (nidx < ne00 && blockIdx.y < ne01 && blockIdx.z < ne02*ne03) { | ~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:17:35: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 17 | if (nidx < ne00 && blockIdx.y < ne01 && blockIdx.z < ne02*ne03) { | ~~~~~~~~~~ ^ ~~~~ 8 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:17:56: warning: comparison of integers of different signs: 'unsigned int' and 'int' [-Wsign-compare] 17 | if (nidx < ne00 && blockIdx.y < ne01 && blockIdx.z < ne02*ne03) { | ~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:17:35: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 17 | if (nidx < ne00 && blockIdx.y < ne01 && blockIdx.z < ne02*ne03) { | ~~~~~~~~~~ ^ ~~~~ 8 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:17:56: warning: comparison of integers of different signs: 'unsigned int' and 'int' [-Wsign-compare] 17 | if (nidx < ne00 && blockIdx.y < ne01 && blockIdx.z < ne02*ne03) { | ~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:17:35: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 17 | if (nidx < ne00 && blockIdx.y < ne01 && blockIdx.z < ne02*ne03) { | ~~~~~~~~~~ ^ ~~~~ 161 warnings generated when compiling for gfx1035. 8 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:17:56: warning: comparison of integers of different signs: 'unsigned int' and 'int' [-Wsign-compare] 17 | if (nidx < ne00 && blockIdx.y < ne01 && blockIdx.z < ne02*ne03) { | ~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:17:35: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 17 | if (nidx < ne00 && blockIdx.y < ne01 && blockIdx.z < ne02*ne03) { | ~~~~~~~~~~ ^ ~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:57:34: warning: unused parameter 'nrows_x' [-Wunused-parameter] 57 | const int ncols_x, const int nrows_x, const int nrows_y, const int nrows_dst) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:417:13: warning: 'break' will never be executed [-Wunreachable-code-break] 417 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ 8 warnings generated when compiling for gfx90a. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:17:56: warning: comparison of integers of different signs: 'unsigned int' and 'int' [-Wsign-compare] 17 | if (nidx < ne00 && blockIdx.y < ne01 && blockIdx.z < ne02*ne03) { | ~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:17:35: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 17 | if (nidx < ne00 && blockIdx.y < ne01 && blockIdx.z < ne02*ne03) { | ~~~~~~~~~~ ^ ~~~~ 8 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:17:56: warning: comparison of integers of different signs: 'unsigned int' and 'int' [-Wsign-compare] 17 | if (nidx < ne00 && blockIdx.y < ne01 && blockIdx.z < ne02*ne03) { | ~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pad.cu:17:35: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 17 | if (nidx < ne00 && blockIdx.y < ne01 && blockIdx.z < ne02*ne03) { | ~~~~~~~~~~ ^ ~~~~ 8 warnings generated when compiling for host. [ 40%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/pool2d.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/pool2d.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/pool2d.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/pool2d.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pool2d.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pool2d.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pool2d.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pool2d.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pool2d.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pool2d.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pool2d.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pool2d.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pool2d.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pool2d.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pool2d.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1035. 161 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pool2d.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pool2d.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:57:34: warning: unused parameter 'nrows_x' [-Wunused-parameter] 57 | const int ncols_x, const int nrows_x, const int nrows_y, const int nrows_dst) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:417:13: warning: 'break' will never be executed [-Wunreachable-code-break] 417 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pool2d.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pool2d.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ 6 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pool2d.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pool2d.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pool2d.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pool2d.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pool2d.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pool2d.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pool2d.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pool2d.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pool2d.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pool2d.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pool2d.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pool2d.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pool2d.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pool2d.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx90a. 161 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pool2d.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pool2d.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx90a. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:57:34: warning: unused parameter 'nrows_x' [-Wunused-parameter] 57 | const int ncols_x, const int nrows_x, const int nrows_y, const int nrows_dst) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:417:13: warning: 'break' will never be executed [-Wunreachable-code-break] 417 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pool2d.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pool2d.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pool2d.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/pool2d.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for host. [ 41%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/quantize.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/quantize.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/quantize.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/quantize.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:4: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:4: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:167:13: warning: 'break' will never be executed [-Wunreachable-code-break] 167 | break; | ^~~~~ 15 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:4: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:4: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:167:13: warning: 'break' will never be executed [-Wunreachable-code-break] 167 | break; | ^~~~~ 15 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:4: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:4: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:167:13: warning: 'break' will never be executed [-Wunreachable-code-break] 167 | break; | ^~~~~ 15 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:4: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:4: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:167:13: warning: 'break' will never be executed [-Wunreachable-code-break] 167 | break; | ^~~~~ 15 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:4: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:4: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:167:13: warning: 'break' will never be executed [-Wunreachable-code-break] 167 | break; | ^~~~~ 15 warnings generated when compiling for gfx1035. 161 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:4: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:4: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:167:13: warning: 'break' will never be executed [-Wunreachable-code-break] 167 | break; | ^~~~~ 15 warnings generated when compiling for gfx1100. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:57:34: warning: unused parameter 'nrows_x' [-Wunused-parameter] 57 | const int ncols_x, const int nrows_x, const int nrows_y, const int nrows_dst) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:417:13: warning: 'break' will never be executed [-Wunreachable-code-break] 417 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:4: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:4: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:167:13: warning: 'break' will never be executed [-Wunreachable-code-break] 167 | break; | ^~~~~ 15 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:4: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:4: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:167:13: warning: 'break' will never be executed [-Wunreachable-code-break] 167 | break; | ^~~~~ 15 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:4: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:4: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:167:13: warning: 'break' will never be executed [-Wunreachable-code-break] 167 | break; | ^~~~~ 15 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:4: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:4: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:167:13: warning: 'break' will never be executed [-Wunreachable-code-break] 167 | break; | ^~~~~ 15 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:4: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:4: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:167:13: warning: 'break' will never be executed [-Wunreachable-code-break] 167 | break; | ^~~~~ 15 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:4: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:4: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:167:13: warning: 'break' will never be executed [-Wunreachable-code-break] 167 | break; | ^~~~~ 15 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:4: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:4: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:167:13: warning: 'break' will never be executed [-Wunreachable-code-break] 167 | break; | ^~~~~ 161 warnings generated when compiling for gfx1103. 15 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:57:34: warning: unused parameter 'nrows_x' [-Wunused-parameter] 57 | const int ncols_x, const int nrows_x, const int nrows_y, const int nrows_dst) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:417:13: warning: 'break' will never be executed [-Wunreachable-code-break] 417 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:4: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:4/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ : /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter]/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ :1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ const int * __restric/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cudat(_vx_, xv,y ,c odnsstt, inncto l*s __x_,r ensrtorwisc_tx_,_ nyr,o wfsl_oya,t nc* o_ls__rey,s tnrriocwts___d sstu,m ,s tcroenasmt) ;i n t| ^& k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:167:13: warning: 'break' will never be executed [-Wunreachable-code-break] 167 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ 15 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:4: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:4: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:167:13: warning: 'break' will never be executed [-Wunreachable-code-break] 167 | break; | ^~~~~ 15 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:4: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:4: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:167:13: warning: 'break' will never be executed [-Wunreachable-code-break] 167 | break; | ^~~~~ 15 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:4: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cuh:4: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/quantize.cu:167:13: warning: 'break' will never be executed [-Wunreachable-code-break] 167 | break; | ^~~~~ 15 warnings generated when compiling for host. [ 42%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/rope.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/rope.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/rope.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/rope.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/rope.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/rope.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/rope.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/rope.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/rope.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/rope.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/rope.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1030. 161 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/rope.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/rope.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:57:34: warning: unused parameter 'nrows_x' [-Wunused-parameter] 57 | const int ncols_x, const int nrows_x, const int nrows_y, const int nrows_dst) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:417:13: warning: 'break' will never be executed [-Wunreachable-code-break] 417 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ 6 warnings generated when compiling for gfx1031. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/rope.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/rope.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/rope.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/rope.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/rope.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/rope.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/rope.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/rope.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/rope.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/rope.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/rope.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/rope.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/rope.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/rope.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/rope.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/rope.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/rope.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/rope.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/rope.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/rope.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/rope.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/rope.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/rope.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/rope.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/rope.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/rope.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for host. [ 43%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/scale.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/scale.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/scale.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/scale.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/scale.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/scale.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/scale.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/scale.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/scale.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/scale.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/scale.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/scale.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/scale.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/scale.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/scale.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/scale.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/scale.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/scale.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/scale.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/scale.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/scale.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 293 warnings generated when compiling for gfx900. 6 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/scale.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/scale.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:57:34: warning: unused parameter 'nrows_x' [-Wunused-parameter] 57 | const int ncols_x, const int nrows_x, const int nrows_y, const int nrows_dst) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:417:13: warning: 'break' will never be executed [-Wunreachable-code-break] 417 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ 6 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/scale.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/scale.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/scale.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/scale.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/scale.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/scale.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/scale.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/scale.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/scale.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/scale.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/scale.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/scale.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/scale.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/scale.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/scale.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/scale.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for host. [ 44%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/softmax.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/softmax.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/softmax.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/softmax.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ 18 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ 18 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ 18 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ 18 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ 18 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ 18 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ 18 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ 18 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ 18 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 293 warnings generated when compiling for gfx906. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ 18 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:57:34: warning: unused parameter 'nrows_x' [-Wunused-parameter] 57 | const int ncols_x, const int nrows_x, const int nrows_y, const int nrows_dst) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:417:13: warning: 'break' will never be executed [-Wunreachable-code-break] 417 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9 :215 | warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] mul _261m | a t _ v e c _ q _csutrdua(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ 18 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ 18 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ 18 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ 18 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ 18 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:37:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 37 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:73:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 73 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:106:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 106 | for (int col0 = 0; col0 < ncols; col0 += block_size) { | ^ 18 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/softmax.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for host. [ 45%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/sum.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/sum.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/sum.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/sum.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sum.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sum.cu:12: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sum.cu:12: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sum.cu:12: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sum.cu:12: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sum.cu:12: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sum.cu:12: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sum.cu:12: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sum.cu:12: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sum.cu:12: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sum.cu:12: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sum.cu:12: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sum.cu:12: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sum.cu:12: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sum.cu:12: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 293 warnings generated when compiling for gfx908. 6 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sum.cu:12: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:57:34: warning: unused parameter 'nrows_x' [-Wunused-parameter] 57 | const int ncols_x, const int nrows_x, const int nrows_y, const int nrows_dst) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:417:13: warning: 'break' will never be executed [-Wunreachable-code-break] 417 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ 6 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sum.cu:12: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sum.cu:12: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for host. [ 46%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/sumrows.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/sumrows.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/sumrows.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/sumrows.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/sumrows.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for host. [ 47%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/tsembd.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/tsembd.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/tsembd.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/tsembd.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/tsembd.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/tsembd.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/tsembd.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/tsembd.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/tsembd.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/tsembd.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/tsembd.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 293 warnings generated when compiling for gfx90a. 6 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:57:34: warning: unused parameter 'nrows_x' [-Wunused-parameter] 57 | const int ncols_x, const int nrows_x, const int nrows_y, const int nrows_dst) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:417:13: warning: 'break' will never be executed [-Wunreachable-code-break] 417 | break; | ^~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/tsembd.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/tsembd.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ 6 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/tsembd.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/tsembd.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/tsembd.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/tsembd.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/tsembd.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/tsembd.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/tsembd.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/tsembd.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/tsembd.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/tsembd.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/tsembd.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/tsembd.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/tsembd.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/tsembd.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/tsembd.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/tsembd.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/tsembd.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/tsembd.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/tsembd.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/tsembd.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/tsembd.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/tsembd.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/tsembd.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/tsembd.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/tsembd.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/tsembd.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for host. [ 48%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/unary.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/unary.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/unary.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/unary.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/unary.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/unary.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/unary.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/unary.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/unary.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/unary.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/unary.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/unary.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/unary.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/unary.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/unary.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/unary.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/unary.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/unary.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/unary.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/unary.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/unary.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 293 warnings generated when compiling for gfx90a. 6 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:57:34: warning: unused parameter 'nrows_x' [-Wunused-parameter] 57 | const int ncols_x, const int nrows_x, const int nrows_y, const int nrows_dst) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:417:13: warning: 'break' will never be executed [-Wunreachable-code-break] 417 | break; | ^~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/unary.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/unary.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ 6 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/unary.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/unary.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/unary.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/unary.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/unary.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/unary.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/unary.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/unary.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/unary.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/unary.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/unary.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/unary.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/unary.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/unary.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/unary.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/unary.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for host. [ 49%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/upscale.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/upscale.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/upscale.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/upscale.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cu:22:37: warning: cast from 'const float *' to 'char *' drops const qualifier [-Wcast-qual] 22 | dst[index] = *(float *)((char *)x + i03 * nb03 + i02 * nb02 + i01 * nb01 + i00 * nb00); | ^ 7 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cu:22:37: warning: cast from 'const float *' to 'char *' drops const qualifier [-Wcast-qual] 22 | dst[index] = *(float *)((char *)x + i03 * nb03 + i02 * nb02 + i01 * nb01 + i00 * nb00); | ^ 7 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cu:22:37: warning: cast from 'const float *' to 'char *' drops const qualifier [-Wcast-qual] 22 | dst[index] = *(float *)((char *)x + i03 * nb03 + i02 * nb02 + i01 * nb01 + i00 * nb00); | ^ 7 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cu:22:37: warning: cast from 'const float *' to 'char *' drops const qualifier [-Wcast-qual] 22 | dst[index] = *(float *)((char *)x + i03 * nb03 + i02 * nb02 + i01 * nb01 + i00 * nb00); | ^ 7 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cu:22:37: warning: cast from 'const float *' to 'char *' drops const qualifier [-Wcast-qual] 22 | dst[index] = *(float *)((char *)x + i03 * nb03 + i02 * nb02 + i01 * nb01 + i00 * nb00); | ^ 7 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cu:22:37: warning: cast from 'const float *' to 'char *' drops const qualifier [-Wcast-qual] 22 | dst[index] = *(float *)((char *)x + i03 * nb03 + i02 * nb02 + i01 * nb01 + i00 * nb00); | ^ 7 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cu:22:37: warning: cast from 'const float *' to 'char *' drops const qualifier [-Wcast-qual] 22 | dst[index] = *(float *)((char *)x + i03 * nb03 + i02 * nb02 + i01 * nb01 + i00 * nb00); | ^ 7 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cu:22:37: warning: cast from 'const float *' to 'char *' drops const qualifier [-Wcast-qual] 22 | dst[index] = *(float *)((char *)x + i03 * nb03 + i02 * nb02 + i01 * nb01 + i00 * nb00); | ^ 7 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cu:22:37: warning: cast from 'const float *' to 'char *' drops const qualifier [-Wcast-qual] 22 | dst[index] = *(float *)((char *)x + i03 * nb03 + i02 * nb02 + i01 * nb01 + i00 * nb00); | ^ 7 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cu:22:37: warning: cast from 'const float *' to 'char *' drops const qualifier [-Wcast-qual] 22 | dst[index] = *(float *)((char *)x + i03 * nb03 + i02 * nb02 + i01 * nb01 + i00 * nb00); | ^ 7 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cu:22:37: warning: cast from 'const float *' to 'char *' drops const qualifier [-Wcast-qual] 22 | dst[index] = *(float *)((char *)x + i03 * nb03 + i02 * nb02 + i01 * nb01 + i00 * nb00); | ^ 7 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cu:22:37: warning: cast from 'const float *' to 'char *' drops const qualifier [-Wcast-qual] 22 | dst[index] = *(float *)((char *)x + i03 * nb03 + i02 * nb02 + i01 * nb01 + i00 * nb00); | ^ 7 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cu:22:37: warning: cast from 'const float *' to 'char *' drops const qualifier [-Wcast-qual] 22 | dst[index] = *(float *)((char *)x + i03 * nb03 + i02 * nb02 + i01 * nb01 + i00 * nb00); | ^ 7 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cu:22:37: warning: cast from 'const float *' to 'char *' drops const qualifier [-Wcast-qual] 22 | dst[index] = *(float *)((char *)x + i03 * nb03 + i02 * nb02 + i01 * nb01 + i00 * nb00); | ^ 7 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cu:22:37: warning: cast from 'const float *' to 'char *' drops const qualifier [-Wcast-qual] 22 | dst[index] = *(float *)((char *)x + i03 * nb03 + i02 * nb02 + i01 * nb01 + i00 * nb00); | ^ 7 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cu:22:37: warning: cast from 'const float *' to 'char *' drops const qualifier [-Wcast-qual] 22 | dst[index] = *(float *)((char *)x + i03 * nb03 + i02 * nb02 + i01 * nb01 + i00 * nb00); | ^ 7 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/upscale.cu:22:37: warning: cast from 'const float *' to 'char *' drops const qualifier [-Wcast-qual] 22 | dst[index] = *(float *)((char *)x + i03 * nb03 + i02 * nb02 + i01 * nb01 + i00 * nb00); | ^ 7 warnings generated when compiling for host. [ 50%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/wkv6.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/wkv6.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/wkv6.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/wkv6.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/wkv6.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/wkv6.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/wkv6.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/wkv6.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1030. 293 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/wkv6.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1031. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:57:34: warning: unused parameter 'nrows_x' [-Wunused-parameter] 57 | const int ncols_x, const int nrows_x, const int nrows_y, const int nrows_dst) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:417:13: warning: 'break' will never be executed [-Wunreachable-code-break] 417 | break; | ^~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:208:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 208 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:215:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 215 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:222:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 222 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:229:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 229 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:236:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 236 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:243:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 243 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:250:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 250 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:257:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 257 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:264:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 264 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:271:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 271 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:278:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 278 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:285:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 285 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:292:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 292 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:299:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 299 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:306:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 306 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:313:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 313 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:320:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 320 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:327:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 327 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:175:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 175 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:178:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 178 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:181:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 181 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:184:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 184 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:187:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 187 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:190:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 190 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:193:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 193 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:80:48: warning: suggest braces around initialization of subobject [-Wmissing-braces] 80 | float tmp[ncols_y][rows_per_cuda_block] = {0.0f}; | ^~~~ | { } /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:196:13: note: in instantiation of function template specialization 'mul_mat_vec_q' requested here 196 | mul_mat_vec_q<<>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:334:5: note: in instantiation of function template specialization 'mul_mat_vec_q_cuda' requested here 334 | mul_mat_vec_q_cuda(vx, vy, dst, ncols_x, nrows_x, nrows_y, ncols_y, nrows_dst, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/mmvq.cu:126:98: warning: comparison of integers of different signs: 'unsigned int' and 'const int' [-Wsign-compare] 126 | if (threadIdx.x < rows_per_cuda_block && (rows_per_cuda_block == 1 || row0 + threadIdx.x < nrows_dst)) { | ~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/wkv6.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1035. 293 warnings generated when compiling for host. [ 51%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/wkv6.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 64 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/wkv6.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/wkv6.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 64 warnings generated when compiling for gfx1012. 6 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/wkv6.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 6 warnings generated when compiling for gfx1103. 64 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/wkv6.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/wkv6.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 64 warnings generated when compiling for gfx1031. 6 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/wkv6.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175 :5059 | : warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] laun c175h | _ f a t t n < D ,s tpraurcatl l{e l _| b ^l ocks>(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 64 warnings generated when compiling for gfx1035. 6 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/wkv6.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 6 warnings generated when compiling for gfx908. 64 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/wkv6.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 6 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/wkv6.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 64 warnings generated when compiling for gfx1101. 6 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/wkv6.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ 3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5:/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h :note: 288in instantiation of function template specialization 'launch_fattn<128, 1>' requested here: 9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 505 | 288 | l a u n c h _ fsattrtunc(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 64 warnings generated when compiling for gfx1102. 6 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/wkv6.cu:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ 6 warnings generated when compiling for host. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ [ 52%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu.o In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 64 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ 61 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 64 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 61 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 64 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 61 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 64 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 61 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 64 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 61 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 64 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 61 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 64 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 61 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 64 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 61 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 64 warnings generated when compiling for host. [ 53%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 61 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 64 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 61 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 64 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 61 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 64 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 61 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 64 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 61 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ 64 warnings generated when compiling for gfx1035. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 61 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 64 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 61 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 64 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 61 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 64 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 61 warnings generated when compiling for host. [ 54%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 64 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 64 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 64 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 64 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 64 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 64 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 64 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 64 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ 64 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 64 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 64 warnings generated when compiling for gfx90a. 64 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results: 9 :| ^ warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here154 | 491 | s t r u c t { l a| u ^n ch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ :3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ : /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 64 warnings generated when compiling for gfx90a. 64 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 64 warnings generated when compiling for gfx942. 64 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 64 warnings generated when compiling for host. [ 55%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu 64 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 58 warnings generated when compiling for gfx1010. 64 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 58 warnings generated when compiling for gfx1012. 64 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 58 warnings generated when compiling for gfx1030. 64 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ 554 | /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 58 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 64 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ 58 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 64 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 58 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 64 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 58 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 64 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 58 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<80, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<80, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<80, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<80, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<112, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<112, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<112, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<112, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 64 warnings generated when compiling for host. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ [ 56%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq1_s.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq1_s.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq1_s.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq1_s.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 58 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.hs:t196r:e9a:m _warning: kanonymous types declared in an anonymous union are an extension [-Wnested-anon-types]_ fixu p196< | t y p e , m m qs_txr,u cMtM Q{_ N W| A ^R PS, need_check><<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 58 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 58 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 58 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 58 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 58 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 58 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ 58 warnings generated when compiling for gfx942. 78 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:14:35: warning: unused parameter 'Q' [-Wunused-parameter] 14 | const char * __restrict__ Q, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:15:35: warning: unused parameter 'K' [-Wunused-parameter] 15 | const char * __restrict__ K, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:16:35: warning: unused parameter 'V' [-Wunused-parameter] 16 | const char * __restrict__ V, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:17:35: warning: unused parameter 'mask' [-Wunused-parameter] 17 | const char * __restrict__ mask, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:18:35: warning: unused parameter 'dst' [-Wunused-parameter] 18 | float * __restrict__ dst, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:19:35: warning: unused parameter 'dst_meta' [-Wunused-parameter] 19 | float2 * __restrict__ dst_meta, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:20:21: warning: unused parameter 'scale' [-Wunused-parameter] 20 | const float scale, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:21:21: warning: unused parameter 'max_bias' [-Wunused-parameter] 21 | const float max_bias, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:22:21: warning: unused parameter 'm0' [-Wunused-parameter] 22 | const float m0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:23:21: warning: unused parameter 'm1' [-Wunused-parameter] 23 | const float m1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:24:24: warning: unused parameter 'n_head_log2' [-Wunused-parameter] 24 | const uint32_t n_head_log2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:25:21: warning: unused parameter 'logit_softcap' [-Wunused-parameter] 25 | const float logit_softcap, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:26:19: warning: unused parameter 'ne00' [-Wunused-parameter] 26 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:27:19: warning: unused parameter 'ne01' [-Wunused-parameter] 27 | const int ne01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:28:19: warning: unused parameter 'ne02' [-Wunused-parameter] 28 | const int ne02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:29:19: warning: unused parameter 'ne03' [-Wunused-parameter] 29 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:30:19: warning: unused parameter 'ne10' [-Wunused-parameter] 30 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:31:19: warning: unused parameter 'ne11' [-Wunused-parameter] 31 | const int ne11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:32:19: warning: unused parameter 'ne12' [-Wunused-parameter] 32 | const int ne12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:33:19: warning: unused parameter 'ne13' [-Wunused-parameter] 33 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:34:19: warning: unused parameter 'ne31' [-Wunused-parameter] 34 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:35:19: warning: unused parameter 'nb31' [-Wunused-parameter] 35 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:36:19: warning: unused parameter 'nb01' [-Wunused-parameter] 36 | const int nb01, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:37:19: warning: unused parameter 'nb02' [-Wunused-parameter] 37 | const int nb02, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:38:19: warning: unused parameter 'nb03' [-Wunused-parameter] 38 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:39:19: warning: unused parameter 'nb11' [-Wunused-parameter] 39 | const int nb11, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:40:19: warning: unused parameter 'nb12' [-Wunused-parameter] 40 | const int nb12, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:41:19: warning: unused parameter 'nb13' [-Wunused-parameter] 41 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:42:19: warning: unused parameter 'nb21' [-Wunused-parameter] 42 | const int nb21, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:43:19: warning: unused parameter 'nb22' [-Wunused-parameter] 43 | const int nb22, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:44:19: warning: unused parameter 'nb23' [-Wunused-parameter] 44 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:45:19: warning: unused parameter 'ne0' [-Wunused-parameter] 45 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:46:19: warning: unused parameter 'ne1' [-Wunused-parameter] 46 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:47:19: warning: unused parameter 'ne2' [-Wunused-parameter] 47 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:48:19: warning: unused parameter 'ne3' [-Wunused-parameter] 48 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<64, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<96, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<96, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<96, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<96, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<128, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:476:9: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 476 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 2>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:491:9: note: in instantiation of function template specialization 'launch_fattn<256, 2>' requested here 491 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-wmma-f16.cuh:505:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 505 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 58 warnings generated when compiling for host. [ 56%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq2_s.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq2_s.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq2_s.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq2_s.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1102. 78 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 78 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq1_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for host. [ 57%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for host. [ 58%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for host. [ 59%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq3_s.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq3_s.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq3_s.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq3_s.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for host. [ 60%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 78 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx906. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_s.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for host. [ 61%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for host. [ 62%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for host. [ 63%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q2_k.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q2_k.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q2_k.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q2_k.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 1022 | for (int k01 = 0; k01 < WARP_SIZE; k01 += QR2_K*VDR_Q2_K_Q8_1_MMQ) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 78 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for host. [ 64%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q3_k.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q3_k.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q3_k.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q3_k.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 98 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 1022 | for (int k01 = 0; k01 < WARP_SIZE; k01 += QR2_K*VDR_Q2_K_Q8_1_MMQ) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 78 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 118 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q3_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for host. [ 65%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q4_0.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q4_0.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q4_0.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q4_0.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for host. [ 66%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q4_1.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q4_1.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q4_1.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q4_1.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 1022 | for (int k01 = 0; k01 < WARP_SIZE; k01 += QR2_K*VDR_Q2_K_Q8_1_MMQ) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 78 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 110 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_1.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for host. [ 67%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q4_k.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q4_k.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q4_k.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q4_k.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx906. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 1022 | for (int k01 = 0; k01 < WARP_SIZE; k01 += QR2_K*VDR_Q2_K_Q8_1_MMQ) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1022:5: warning: loop not unrolled: the optimizer was unable to perform the requested transformation; the transformation might be disabled or specified as part of an unsupported transformation ordering [-Wpass-failed=transform-warning] 78 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 134 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q2_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for host. [ 68%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q5_0.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q5_0.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q5_0.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q5_0.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q4_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for host. [ 69%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q5_1.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q5_1.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q5_1.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q5_1.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1035. 78 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare]> > | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870: 132691: | note: in instantiation of function template specialization 'launch_mul_mat_q' requested here i2870f | ( i t ! = b l o c klIaduxn.cxh _|m|u lj_tm a!t=_ qb ({c t x| , ~~ ^ ~~~~~~~~~~ args, stream); | ^/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh :2805:9:/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh :note: 2691in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here: 16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2805 | 2691 | m u li_fm a(ti_tq _!s=t rbelaomc_kkI_dfxi.xxu p|<|t yjpte ,! =m mbql_oxc,k IMdMxQ._yN)W A{R P S| , ~~ ^ ~~~~~~~~~~ need_check><<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt !/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ ims, 0, stream>>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != b/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9lockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | : note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2805 | 2885 | m u l _ m a tl_aqu_nscthr_emauml__km_afti_xquq(_cxt,x ,M MaQr_gNsW,A RsPtSr,e anme)e;d _ c| h ^e ck><<k>I>d x .| x ^ || jt != b/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuhl:o2870c:k13I:d xnote: .in instantiation of function template specialization 'launch_mul_mat_q' requested herey ) { | ~~ ^ ~~~~~~~~~~2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup >> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_m/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jtat_ != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 2>(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for host. [ 70%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q5_k.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q5_k.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q5_k.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q5_k.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_1.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for host. [ 71%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q6_k.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q6_k.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q6_k.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q6_k.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx900. 78 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx906. 78 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9:/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q5_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for host. [ 72%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q8_0.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q8_0.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q8_0.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q8_0.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q6_k.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for host. [ 73%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:334:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 334 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:344:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 344 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:347:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 347 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:357:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 357 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:360:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:370:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 370 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:373:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 373 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:385:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 385 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx1010. 78 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:334:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 334 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:344:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 344 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:347:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 347 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:357:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 357 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:360:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:370:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 370 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:373:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 373 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:385:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 385 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 60 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:334:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 334 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:344:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 344 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:347:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 347 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:357:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 357 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:360:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:370:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 370 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:373:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 373 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:385:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 385 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 78 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 60 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:334:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 334 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:344:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 344 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:347:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 347 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:357:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 357 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:360:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:370:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 370 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:373:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 373 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:385:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 385 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:334:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 334 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:344:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 344 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:347:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 347 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:357:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 357 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:360:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:370:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 370 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:373:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 373 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:385:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 385 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 78 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 60 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:334:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 334 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:344:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 344 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:347:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 347 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:357:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 357 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:360:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:370:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 370 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:373:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 373 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:385:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 385 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:334:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 334 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:344:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 344 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:347:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 347 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:357:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 357 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:360:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:370:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 370 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:373:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 373 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:385:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 385 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:334:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 334 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:344:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 344 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:347:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 347 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:357:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 357 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:360:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:370:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 370 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:373:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 373 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:385:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 385 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:334:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 334 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:344:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 344 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:347:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 347 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:357:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 357 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:360:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:370:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 370 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:373:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 373 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:385:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 385 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:334:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 334 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:344:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 344 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:347:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 347 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:357:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 357 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:360:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:370:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 370 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:373:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 373 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:385:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 385 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 78 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:172:105: warning: function 'mma_K4' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 172 | __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mma.cuh:194:105: warning: function 'mma_K8' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 194 | __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/mmq-instance-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:866:99: warning: unused parameter 'k00' [-Wunused-parameter] 866 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1053:99: warning: unused parameter 'k00' [-Wunused-parameter] 1053 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:1704:99: warning: unused parameter 'k00' [-Wunused-parameter] 1704 | const int * __restrict__ x, const int * __restrict__ y, float * __restrict__ sum, const int & k00) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:17: warning: unused parameter 'ne00' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2497:75: warning: unused parameter 'ne10' [-Wunused-parameter] 2497 | const int & ne00, const int & ne01, const int & stride01, const int & ne10, const int & ne11, const int & stride11, const int & ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2821:15: warning: unused variable 'nsm' [-Wunused-variable] 2821 | const int nsm = ggml_cuda_info().devices[id].nsm; | ^~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2852:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2852 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2855:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2855 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2858:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2858 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2861:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2861 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2864:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2864 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2867:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2867 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2870:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2870 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2873:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2873 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2876:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2876 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2879:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2879 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2882:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2882 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2885:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2885 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2888:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2888 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2891:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2891 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2894:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2894 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2805:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2805 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:36: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2813:9: note: in instantiation of function template specialization 'mul_mat_q_stream_k_fixup' requested here 2813 | mul_mat_q_stream_k_fixup<<>> | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2897:13: note: in instantiation of function template specialization 'launch_mul_mat_q' requested here 2897 | launch_mul_mat_q(ctx, args, stream); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../mmq.cuh:2691:16: warning: comparison of integers of different signs: 'const int' and 'unsigned int' [-Wsign-compare] 2691 | if (it != blockIdx.x || jt != blockIdx.y) { | ~~ ^ ~~~~~~~~~~ 78 warnings generated when compiling for host. [ 74%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:312:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 312 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:322:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 322 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:325:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 325 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:335:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 335 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:338:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 338 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:348:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 348 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:351:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 351 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:363:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 363 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:334:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 334 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:344:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 344 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:347:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 347 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:357:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 357 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:360:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:370:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 370 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:373:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 373 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:385:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 385 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx1010. 60 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:312:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 312 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:322:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 322 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:325:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 325 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:335:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 335 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:338:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 338 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:348:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 348 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:351:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 351 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:363:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 363 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:334:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 334 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:344:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 344 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:347:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 347 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:357:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 357 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:360:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:370:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 370 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:373:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 373 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:385:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 385 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx906. 60 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:334:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 334 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:344:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 344 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:347:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 347 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:357:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 357 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:360:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:370:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 370 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:373:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 373 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:385:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 385 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:312:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 312 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:322:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 322 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:325:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 325 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:335:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 335 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:338:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 338 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:348:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 348 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:351:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 351 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:363:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 363 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:334:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 334 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:344:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 344 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:347:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 347 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:357:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 357 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:360:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:370:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 370 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:373:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 373 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:385:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 385 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:312:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 312 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:322:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 322 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:325:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 325 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:335:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 335 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:338:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 338 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:348:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 348 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:351:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 351 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:363:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 363 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:334:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 334 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:344:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 344 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:347:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 347 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:357:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 357 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:360:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:370:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 370 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:373:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 373 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:385:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 385 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:312:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 312 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:322:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 322 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:325:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 325 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:335:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 335 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:338:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 338 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:348:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 348 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:351:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 351 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:363:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 363 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:334:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 334 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:344:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 344 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:347:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 347 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:357:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 357 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:360:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:370:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 370 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:373:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 373 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:385:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 385 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:312:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 312 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:322:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 322 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:325:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 325 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:335:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 335 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:338:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 338 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:348:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 348 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:351:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 351 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:363:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 363 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:334:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 334 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:344:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 344 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:347:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 347 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:357:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 357 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:360:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:370:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 370 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:373:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 373 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:385:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 385 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for host. [ 75%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:334:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 334 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:344:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 344 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:347:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 347 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:357:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 357 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:360:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:370:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 370 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:373:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 373 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:385:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 385 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:312:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 312 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:322:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 322 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:325:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 325 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:335:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 335 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:338:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 338 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:348:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 348 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:351:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 351 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:363:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 363 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:334:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 334 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:344:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 344 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:347:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 347 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:357:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 357 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:360:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:370:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 370 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:373:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 373 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:385:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 385 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx1101. 60 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:312:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 312 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33:In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared < h a l ff2o>r( Q(_ifn t+ i40* i=0 ,0 ;s cia0l e<, Dt/mspi_zqe_oif3(2i,n tt)m;p _iq0_ d+s=) ;W A R| P ^_ SIZE)/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh :{304 : 35| : ~~ ^ ~~~~~~~~~~~~~ note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh :304285 | : 35 : note: fin instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested herea ttn_kernel_t 285f | a t t n _fkaetrtnne_lk e=r nfella_sth _fatattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:322:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 322 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ tn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:325:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 325 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_S/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ IZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:335:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 335 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:338:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 338 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:348:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 348 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:351:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 351 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:334:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 334 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:363:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 363 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:344:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 344 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:347:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 347 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:357:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 357 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:360:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:370:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 370 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:373:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 373 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:385:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 385 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:334:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 334 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:344:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 344 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:347:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 347 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:357:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 357 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:360:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:370:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 370 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:373:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 373 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:385:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 385 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:312:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 312 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:322:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 322 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:325:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 325 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:335:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 335 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:338:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 338 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:348:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 348 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:351:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 351 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:363:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 363 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:334:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 334 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:344:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 344 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:347:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 347 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:357:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 357 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:360:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:370:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 370 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:373:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 373 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:385:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 385 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:312:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 312 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:322:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 322 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:325:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 325 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:335:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 335 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:338:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 338 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:348:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 348 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:351:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 351 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:363:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 363 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:334:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 334 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:344:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 344 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:347:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 347 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:357:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 357 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:360:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:370:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 370 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:373:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 373 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:385:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 385 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx1100. 60 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:334:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 334 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:344:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 344 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:347:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 347 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:357:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 357 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:360:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:370:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 370 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:373:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 373 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:385:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 385 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:312:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 312 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:322:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 322 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:325:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 325 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:335:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 335 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:338:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 338 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:348:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 348 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:351:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 351 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:363:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 363 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:334:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 334 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:344:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 344 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:347:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 347 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:357:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 357 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:360:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:370:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 370 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare]60 warnings generated when compiling for gfx900. 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:373:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 373 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:385:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 385 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:312:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 312 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:322:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 322 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:325:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 325 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:335:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 335 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:338:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 338 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:348:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 348 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:351:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 351 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:363:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 363 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:334:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 334 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:344:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 344 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:347:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 347 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:357:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 357 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:360:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:370:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 370 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:373:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 373 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:385:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 385 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:312:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 312 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:322:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 322 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:325:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 325 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:335:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 335 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:338:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 338 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:348:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 348 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:351:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 351 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:363:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 363 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:334:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 334 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:344:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 344 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:347:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 347 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:357:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 357 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:360:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:370:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 370 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:373:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 373 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:385:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 385 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:312:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 312 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:322:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 322 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:325:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 325 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:335:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 335 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:338:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 338 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:348:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 348 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:351:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 351 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:363:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 363 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:334:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 334 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:344:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 344 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:347:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 347 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:357:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 357 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:360:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:370:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 370 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:373:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 373 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:385:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 385 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:334:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 334 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:344:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 344 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:347:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 347 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:357:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 357 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:360:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:370:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 370 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:373:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 373 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:385:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 385 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:312:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 312 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:322:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 322 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:325:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 325 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:335:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 335 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:338:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 338 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:348:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 348 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:351:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 351 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:363:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 363 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:334:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 334 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:344:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 344 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:347:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 347 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:357:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 357 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:360:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:370:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 370 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:373:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 373 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:385:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 385 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:312:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 312 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:322:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 322 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:325:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 325 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:335:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 335 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:338:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 338 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:348:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 348 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:351:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 351 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:363:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 363 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:334:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 334 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:344:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 344 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:347:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 347 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:357:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 357 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:360:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:370:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 370 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:373:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 373 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:385:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 385 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 60 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:334:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 334 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:344:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 344 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:347:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 347 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:357:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 357 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:360:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:370:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 370 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:373:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 373 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:385:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 385 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:312:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 312 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:322:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 322 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:325:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 325 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:335:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 335 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:338:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 338 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:348:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 348 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:351:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 351 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:363:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0, true>' requested here 363 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for host. [ 76%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:312:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 312 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:322:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 322 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:325:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 325 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:335:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 335 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:338:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 338 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:348:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 348 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:351:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 351 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:363:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 363 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:334:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 334 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:344:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 344 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:347:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 347 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:357:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 357 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:360:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:370:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 370 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:373:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 373 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:385:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 385 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:312:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 312 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:322:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 322 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:325:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 325 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:335:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 335 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:338:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 338 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:348:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 348 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:351:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 351 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:363:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 363 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:334:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 334 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:344:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 344 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:347:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 347 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:357:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 357 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:360:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:370:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 370 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:373:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 373 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:385:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 385 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 60 warnings generated when compiling for host. [ 77%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 60 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:312:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 312 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:322:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 322 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:325:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 325 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:335:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 335 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:338:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 338 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:348:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 348 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:351:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 351 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:363:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 363 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 34 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 60 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:312:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 312 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:322:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 322 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:325:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 325 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:335:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 335 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:338:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 338 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:348:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 348 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:351:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 351 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:363:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 363 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 34 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 60 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:312:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 312 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:322:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 322 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:325:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 325 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:335:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 335 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:338:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 338 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:348:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 348 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:351:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 351 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:363:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 363 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 34 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 60 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:312:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 312 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:322:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 322 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:325:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 325 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:335:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 335 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:338:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 338 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:348:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 348 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:351:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 351 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:363:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 363 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 34 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 60 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 34 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:312:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 312 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:322:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 322 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:325:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 325 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:335:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 335 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:338:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 338 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:348:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 348 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:351:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 351 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:363:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 363 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 60 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:312:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 312 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:322:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 322 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:325:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 325 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:335:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 335 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:338:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 338 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:348:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 348 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:351:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 351 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:363:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 363 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 34 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 60 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:312:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 312 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:322:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 322 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:325:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 325 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:335:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 335 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:338:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 338 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:348:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 348 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:351:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 351 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:363:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 363 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 34 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 60 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:312:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 312 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:322:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 322 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:325:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 325 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:335:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 335 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:338:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 338 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:348:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 348 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:351:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 351 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:363:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 363 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 34 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 60 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:312:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 312 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:322:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 322 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:325:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 325 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:335:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 335 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:338:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 338 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:348:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 348 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:351:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 351 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:363:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 363 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 34 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 60 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:312:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 312 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:322:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 322 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:325:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 325 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:335:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 335 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:338:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 338 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:348:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 348 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:351:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 351 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:363:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 363 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 34 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 60 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 34 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:312:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 312 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:322:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 322 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:325:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 325 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:335:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 335 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:338:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 338 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:348:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 348 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:351:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 351 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:363:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 363 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 60 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:312:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 312 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:322:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 322 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:325:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 325 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:335:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 335 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:338:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 338 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:348:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 348 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:351:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 351 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:363:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 363 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 34 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 60 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:312:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 312 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:322:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 322 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:325:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 325 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:335:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 335 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:338:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 338 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:348:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 348 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:351:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 351 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:363:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 363 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 34 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 60 warnings generated when compiling for gfx90a. 34 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:312:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 312 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:322:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 322 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:325:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 325 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:335:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 335 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:338:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 338 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:348:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 348 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:351:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 351 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:363:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 363 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 128>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 128>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 128>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 128>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 128>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<128, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for host. [ 78%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 256>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 256>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 256>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 256>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 256>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 60 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:312:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 312 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:322:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 322 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:325:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 2, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 325 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:335:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 335 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:338:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 4, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 338 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:348:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 348 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:351:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 4, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 351 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:116:37: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 116 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:363:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0, true>' requested here 363 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:129:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 129 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:142:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 142 | for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) { | ~~ ^ ~~~~~~~~~~~~~ 34 warnings generated when compiling for gfx1010. 60 warnings generated when compiling for host. [ 79%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ : warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 256>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 256>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 256>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 256>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 256>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 64>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 64>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 64>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 64>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 64>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 64>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 64>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 64>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 64>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 64>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx1012. 34 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 256>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 256>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 256>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 256>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 256>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 64>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 64>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 64>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 64>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 64>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 64>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 64>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 64>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 64>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 64>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx1030. 34 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 256>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 256>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 256>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 256>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 256>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 64>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 64>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 64>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 64>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 64>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 64>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 64>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 64>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 64>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 64>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx1031. 34 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 256>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 256>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 256>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 256>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 256>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 64>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 64>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 64>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 64>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 64>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 64>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 64>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 64>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 64>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 64>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct 34 warnings generated when compiling for gfx1102. { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 256>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 256>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 256>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 256>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 256>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 64>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 64>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 64>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 64>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 64>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 64>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 64>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 64>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 64>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 64>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ 34 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 256>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 256>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 256>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 256>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 256>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 64>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 64>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 64>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 64>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 64>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 64>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 64>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 64>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 64>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 64>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx1101. 34 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 256>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 256>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 256>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 256>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 256>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 64>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 64>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 64>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 64>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 64>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 64>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 64>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 64>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 64>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 64>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx90a. 34 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 64>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 64>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 64>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 64>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 64>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 256>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 256>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 256>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 256>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 256>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 64>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 64>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 64>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 64>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 64>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx942. 34 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 64>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 64>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 64>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 64>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 64>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<64, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ 34 warnings generated when compiling for host. [ 80%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 256>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 256>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 256>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 256>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 256>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 256>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 256>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 256>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 256>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 256>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 256>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 256>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 256>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 256>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 256>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 34 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 256>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 256>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 256>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 256>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 256>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx908. 34 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 256>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 256>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 256>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 256>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 256>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx1035. 34 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 256>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 256>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 256>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 256>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 256>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx1100. 34 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 256>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 256>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 256>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 256>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 256>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx942. 34 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:483:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0<__half, 256>' requested here 483 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:482:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1<__half, 256>' requested here 482 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (iIn file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb3nt k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:481:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0<__half, 256>' requested here 481 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:480:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1<__half, 256>' requested here 480 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:479:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0<__half, 256>' requested here 479 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared<__half2>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:304:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f16<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 304 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:331:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 331 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:307:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 307 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f16.cuh:382:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f16_case_impl<256, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 382 | ggml_cuda_flash_attn_ext_vec_f16_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for host. [ 81%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx906. 34 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<128, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<128, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<128, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for host. [ 82%] Building CXX object ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/hipcc -DGGML_BACKEND_BUILD -DGGML_BACKEND_SHARED -DGGML_CUDA_DMMV_X=32 -DGGML_CUDA_MMV_Y=1 -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_HIP -DK_QUANTS_PER_ITERATION=2 -DUSE_PROF_API=1 -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -D__HIP_PLATFORM_AMD__=1 -Dggml_hip_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/.. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -x hip --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 -MD -MT ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu.o -MF CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu.o.d -o CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ 34 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx1010. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx1012. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx1030. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx1031. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx1151. 34 warnings generated when compiling for gfx1035. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx1100. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx1101. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx1102. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx1103. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx1151. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx900. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx908. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<64, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<64, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<64, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<64, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for host. 34 warnings generated when compiling for gfx90a. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for gfx942. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../common.cuh:20: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:154:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 154 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:175:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 175 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:196:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 196 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:261:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 261 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:288:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 288 | struct { | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-hip/../ggml-common.h:305:9: warning: anonymous types declared in an anonymous union are an extension [-Wnested-anon-types] 305 | struct { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:563:47: warning: function 'on_no_fattn_vec_case' could be declared with attribute 'noreturn' [-Wmissing-noreturn] 563 | static void on_no_fattn_vec_case(const int D) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:21:19: warning: unused parameter 'ne00' [-Wunused-parameter] 21 | const int ne00, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:24:19: warning: unused parameter 'ne03' [-Wunused-parameter] 24 | const int ne03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:25:19: warning: unused parameter 'ne10' [-Wunused-parameter] 25 | const int ne10, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:28:19: warning: unused parameter 'ne13' [-Wunused-parameter] 28 | const int ne13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:29:19: warning: unused parameter 'ne31' [-Wunused-parameter] 29 | const int ne31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:30:19: warning: unused parameter 'nb31' [-Wunused-parameter] 30 | const int nb31, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:33:19: warning: unused parameter 'nb03' [-Wunused-parameter] 33 | const int nb03, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:36:19: warning: unused parameter 'nb13' [-Wunused-parameter] 36 | const int nb13, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:39:19: warning: unused parameter 'nb23' [-Wunused-parameter] 39 | const int nb23, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:40:19: warning: unused parameter 'ne0' [-Wunused-parameter] 40 | const int ne0, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:41:19: warning: unused parameter 'ne1' [-Wunused-parameter] 41 | const int ne1, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:42:19: warning: unused parameter 'ne2' [-Wunused-parameter] 42 | const int ne2, | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:43:19: warning: unused parameter 'ne3' [-Wunused-parameter] 43 | const int ne3) { | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:247:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 247 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:494:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q8_0' requested here 494 | type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:196:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 196 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:493:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_1' requested here 493 | type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:149:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 149 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:492:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q5_0' requested here 492 | type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:105:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 105 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:491:36: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_1' requested here 491 | type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:65:33: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 65 | for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) { | ~~~~~~ ^ ~~~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:490:39: note: in instantiation of function template specialization 'vec_dot_fattn_vec_KQ_q4_0' requested here 490 | return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0 : | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:318:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 318 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:130:17: note: in instantiation of function template specialization 'quantize_q8_1_to_shared>' requested here 130 | quantize_q8_1_to_shared(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:285:35: note: in instantiation of function template specialization 'flash_attn_vec_ext_f32<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 285 | fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:325:23: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 325 | for (int l = 1; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:341:27: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare] 341 | for (int l = 0; l < sizeof(int); ++l) { | ~ ^ ~~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 4>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<256, 4>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:309:13: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 1, 4, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 309 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu:3: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:554:24: warning: cast from 'const float *' to 'unsigned int *' drops const qualifier [-Wcast-qual] 554 | *((uint32_t *) &KQ_max_scale) &= ftz_mask; | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-common.cuh:704:5: note: in instantiation of function template specialization 'flash_attn_combine_results<256, 1>' requested here 704 | flash_attn_combine_results | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:288:5: note: in instantiation of function template specialization 'launch_fattn<256, 1>' requested here 288 | launch_fattn(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V); | ^ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-cuda/template-instances/../fattn-vec-f32.cuh:360:9: note: in instantiation of function template specialization 'ggml_cuda_flash_attn_ext_vec_f32_case_impl<256, 8, 1, GGML_TYPE_F16, GGML_TYPE_F16, false>' requested here 360 | ggml_cuda_flash_attn_ext_vec_f32_case_impl(ctx, dst); | ^ 34 warnings generated when compiling for host. [ 83%] Linking CXX shared library libggml-hip.so cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/ggml-hip && /usr/bin/cmake -E cmake_link_script CMakeFiles/ggml-hip.dir/link.txt --verbose=1 /usr/bin/hipcc -fPIC -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -Xlinker --dependency-file=CMakeFiles/ggml-hip.dir/link.d -Wl,-z,relro -Wl,--as-needed -Wl,-z,pack-relative-relocs -Wl,-z,now -Wl,-z,now -Wl,--build-id=sha1 -specs=/usr/lib/rpm/redhat/redhat-package-notes -shared -Wl,-soname,libggml-hip.so -o libggml-hip.so "CMakeFiles/ggml-hip.dir/__/ggml-cuda/acc.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/arange.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/argmax.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/argsort.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/binbcast.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/clamp.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/concat.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/conv-transpose-1d.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/convert.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/count-equal.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/cpy.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/cross-entropy-loss.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/diagmask.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/dmmv.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/fattn-tile-f16.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/fattn-tile-f32.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/fattn.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/getrows.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/ggml-cuda.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/im2col.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/mmq.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/mmvq.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/norm.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/opt-step-adamw.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/out-prod.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/pad.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/pool2d.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/quantize.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/rope.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/scale.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/softmax.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/sum.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/sumrows.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/tsembd.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/unary.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/upscale.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/wkv6.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq1_s.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq2_s.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq3_s.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q2_k.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q3_k.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q4_0.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instanclang++: warning: argument unused during compilation: '-Xarch_host -fstack-protector-strong' [-Wunused-command-line-argument] clang++: warning: argument unused during compilation: '-Xarch_host -fcf-protection' [-Wunused-command-line-argument] clang++: warning: argument unused during compilation: '-specs=/usr/lib/rpm/redhat/redhat-package-notes' [-Wunused-command-line-argument] ces/mmq-instance-q4_1.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q4_k.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q5_0.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q5_1.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q5_k.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q6_k.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/mmq-instance-q8_0.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q8_0-q8_0.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu.o" "CMakeFiles/ggml-hip.dir/__/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu.o" ../libggml-base.so.b4094 /usr/lib64/libhipblas.so.2.3 --hip-link --offload-arch=gfx900 --offload-arch=gfx906:xnack- --offload-arch=gfx908:xnack- --offload-arch=gfx90a:xnack+ --offload-arch=gfx90a:xnack- --offload-arch=gfx942 --offload-arch=gfx1010 --offload-arch=gfx1012 --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1035 --offload-arch=gfx1100 --offload-arch=gfx1101 --offload-arch=gfx1102 --offload-arch=gfx1103 --offload-arch=gfx1151 /usr/lib64/librocblas.so.4.3 /usr/lib64/libamdhip64.so.6.3.42133 gmake[2]: Leaving directory '/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build' [ 83%] Built target ggml-hip /usr/bin/gmake -f ggml/src/CMakeFiles/ggml.dir/build.make ggml/src/CMakeFiles/ggml.dir/depend gmake[2]: Entering directory '/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build' cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build && /usr/bin/cmake -E cmake_depends "Unix Makefiles" /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094 /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src/CMakeFiles/ggml.dir/DependInfo.cmake "--color=" gmake[2]: Leaving directory '/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build' /usr/bin/gmake -f ggml/src/CMakeFiles/ggml.dir/build.make ggml/src/CMakeFiles/ggml.dir/build gmake[2]: Entering directory '/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build' [ 84%] Building CXX object ggml/src/CMakeFiles/ggml.dir/ggml-backend-reg.cpp.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src && /usr/bin/hipcc -DGGML_BACKEND_SHARED -DGGML_BUILD -DGGML_SCHED_MAX_COPIES=4 -DGGML_SHARED -DGGML_USE_CPU -DGGML_USE_CUDA -DGGML_USE_HIP -D_GNU_SOURCE -D_XOPEN_SOURCE=600 -Dggml_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -std=gnu++11 -fPIC -Wmissing-declarations -Wmissing-noreturn -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wunreachable-code-break -Wunreachable-code-return -Wmissing-prototypes -Wextra-semi -MD -MT ggml/src/CMakeFiles/ggml.dir/ggml-backend-reg.cpp.o -MF CMakeFiles/ggml.dir/ggml-backend-reg.cpp.o.d -o CMakeFiles/ggml.dir/ggml-backend-reg.cpp.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/ggml-backend-reg.cpp sh: line 1: /usr/bin/rocm_agent_enumerator: No such file or directory [ 85%] Linking CXX shared library libggml.so cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src && /usr/bin/cmake -E cmake_link_script CMakeFiles/ggml.dir/link.txt --verbose=1 sh: line 1: /usr/bin/rocm_agent_enumerator: No such file or directory clang++: warning: argument unused during compilation: '-Xarch_host -fstack-protector-strong' [-Wunused-command-line-argument] clang++: warning: argument unused during compilation: '-Xarch_host -fcf-protection' [-Wunused-command-line-argument] clang++: warning: argument unused during compilation: '-specs=/usr/lib/rpm/redhat/redhat-package-notes' [-Wunused-command-line-argument] /usr/bin/hipcc -fPIC -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -Xlinker --dependency-file=CMakeFiles/ggml.dir/link.d -Wl,-z,relro -Wl,--as-needed -Wl,-z,pack-relative-relocs -Wl,-z,now -Wl,-z,now -Wl,--build-id=sha1 -specs=/usr/lib/rpm/redhat/redhat-package-notes -shared -Wl,-soname,libggml.so.b4094 -o libggml.so.b4094 "CMakeFiles/ggml.dir/ggml-backend-reg.cpp.o" ggml-cpu/libggml-cpu.so ggml-hip/libggml-hip.so libggml-base.so.b4094 cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/ggml/src && /usr/bin/cmake -E cmake_symlink_library libggml.so.b4094 libggml.so.b4094 libggml.so gmake[2]: Leaving directory '/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build' [ 85%] Built target ggml /usr/bin/gmake -f src/CMakeFiles/llama.dir/build.make src/CMakeFiles/llama.dir/depend gmake[2]: Entering directory '/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build' cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build && /usr/bin/cmake -E cmake_depends "Unix Makefiles" /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094 /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/src /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/src/CMakeFiles/llama.dir/DependInfo.cmake "--color=" gmake[2]: Leaving directory '/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build' /usr/bin/gmake -f src/CMakeFiles/llama.dir/build.make src/CMakeFiles/llama.dir/build gmake[2]: Entering directory '/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build' [ 86%] Building CXX object src/CMakeFiles/llama.dir/llama-vocab.cpp.o [ 87%] Building CXX object src/CMakeFiles/llama.dir/llama.cpp.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/src && /usr/bin/hipcc -DGGML_BACKEND_SHARED -DGGML_SHARED -DGGML_USE_CPU -DGGML_USE_CUDA -DGGML_USE_HIP -DLLAMA_BUILD -DLLAMA_SHARED -Dllama_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/../include -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -fPIC -MD -MT src/CMakeFiles/llama.dir/llama-vocab.cpp.o -MF CMakeFiles/llama.dir/llama-vocab.cpp.o.d -o CMakeFiles/llama.dir/llama-vocab.cpp.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/llama-vocab.cpp cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/src && /usr/bin/hipcc -DGGML_BACKEND_SHARED -DGGML_SHARED -DGGML_USE_CPU -DGGML_USE_CUDA -DGGML_USE_HIP -DLLAMA_BUILD -DLLAMA_SHARED -Dllama_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/../include -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -fPIC -MD -MT src/CMakeFiles/llama.dir/llama.cpp.o -MF CMakeFiles/llama.dir/llama.cpp.o.d -o CMakeFiles/llama.dir/llama.cpp.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/llama.cpp sh: line 1: /usr/bin/rocm_agent_enumerator: No such file or directory sh: line 1: /usr/bin/rocm_agent_enumerator: No such file or directory /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/llama-vocab.cpp:114:13: warning: unused function 'llama_is_unknown_token' [-Wunused-function] 114 | static bool llama_is_unknown_token(const llama_vocab & vocab, llama_token id) { | ^~~~~~~~~~~~~~~~~~~~~~ 1 warning generated when compiling for gfx906. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/llama-vocab.cpp:114:13: warning: unused function 'llama_is_unknown_token' [-Wunused-function] 114 | static bool llama_is_unknown_token(const llama_vocab & vocab, llama_token id) { | ^~~~~~~~~~~~~~~~~~~~~~ 1 warning generated when compiling for host. [ 88%] Building CXX object src/CMakeFiles/llama.dir/llama-grammar.cpp.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/src && /usr/bin/hipcc -DGGML_BACKEND_SHARED -DGGML_SHARED -DGGML_USE_CPU -DGGML_USE_CUDA -DGGML_USE_HIP -DLLAMA_BUILD -DLLAMA_SHARED -Dllama_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/../include -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -fPIC -MD -MT src/CMakeFiles/llama.dir/llama-grammar.cpp.o -MF CMakeFiles/llama.dir/llama-grammar.cpp.o.d -o CMakeFiles/llama.dir/llama-grammar.cpp.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/llama-grammar.cpp sh: line 1: /usr/bin/rocm_agent_enumerator: No such file or directory In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/llama-grammar.cpp:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/llama-grammar.h:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/llama-impl.h:52:13: warning: unused function 'replace_all' [-Wunused-function] 52 | static void replace_all(std::string & s, const std::string & search, const std::string & replace) { | ^~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/llama-grammar.cpp:201:13: warning: unused function 'print_rule_binary' [-Wunused-function] 201 | static void print_rule_binary(FILE * file, const llama_grammar_rule & rule) { | ^~~~~~~~~~~~~~~~~ 2 warnings generated when compiling for gfx906. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/llama.cpp:16701:13: warning: unused function 'llama_set_s_copy' [-Wunused-function] 16701 | static void llama_set_s_copy(llama_context & lctx) { | ^~~~~~~~~~~~~~~~ 1 warning generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/llama-grammar.cpp:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/llama-grammar.h:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/llama-impl.h:52:13: warning: unused function 'replace_all' [-Wunused-function] 52 | static void replace_all(std::string & s, const std::string & search, const std::string & replace) { | ^~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/llama-grammar.cpp:201:13: warning: unused function 'print_rule_binary' [-Wunused-function] 201 | static void print_rule_binary(FILE * file, const llama_grammar_rule & rule) { | ^~~~~~~~~~~~~~~~~ 2 warnings generated when compiling for host. [ 89%] Building CXX object src/CMakeFiles/llama.dir/llama-sampling.cpp.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/src && /usr/bin/hipcc -DGGML_BACKEND_SHARED -DGGML_SHARED -DGGML_USE_CPU -DGGML_USE_CUDA -DGGML_USE_HIP -DLLAMA_BUILD -DLLAMA_SHARED -Dllama_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/../include -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -fPIC -MD -MT src/CMakeFiles/llama.dir/llama-sampling.cpp.o -MF CMakeFiles/llama.dir/llama-sampling.cpp.o.d -o CMakeFiles/llama.dir/llama-sampling.cpp.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/llama-sampling.cpp sh: line 1: /usr/bin/rocm_agent_enumerator: No such file or directory In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/llama-sampling.cpp:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/llama-sampling.h:5: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/llama-grammar.h:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/llama-impl.h:52:13: warning: unused function 'replace_all' [-Wunused-function] 52 | static void replace_all(std::string & s, const std::string & search, const std::string & replace) { | ^~~~~~~~~~~ 1 warning generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/llama-sampling.cpp:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/llama-sampling.h:5: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/llama-grammar.h:3: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/llama-impl.h:52:13: warning: unused function 'replace_all' [-Wunused-function] 52 | static void replace_all(std::string & s, const std::string & search, const std::string & replace) { | ^~~~~~~~~~~ /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/llama.cpp:16701:13: warning: unused function 'llama_set_s_copy' [-Wunused-function] 16701 | static void llama_set_s_copy(llama_context & lctx) { | ^~~~~~~~~~~~~~~~ 1 warning generated when compiling for host. [ 90%] Building CXX object src/CMakeFiles/llama.dir/unicode.cpp.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/src && /usr/bin/hipcc -DGGML_BACKEND_SHARED -DGGML_SHARED -DGGML_USE_CPU -DGGML_USE_CUDA -DGGML_USE_HIP -DLLAMA_BUILD -DLLAMA_SHARED -Dllama_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/../include -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -fPIC -MD -MT src/CMakeFiles/llama.dir/unicode.cpp.o -MF CMakeFiles/llama.dir/unicode.cpp.o.d -o CMakeFiles/llama.dir/unicode.cpp.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/unicode.cpp sh: line 1: /usr/bin/rocm_agent_enumerator: No such file or directory /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/unicode.cpp:29:20: warning: unused function 'unicode_cpts_to_utf8' [-Wunused-function] 29 | static std::string unicode_cpts_to_utf8(const std::vector & cps) { | ^~~~~~~~~~~~~~~~~~~~ 1 warning generated when compiling for gfx906. /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/unicode.cpp:29:20: warning: unused function 'unicode_cpts_to_utf8' [-Wunused-function] 29 | static std::string unicode_cpts_to_utf8(const std::vector & cps) { | ^~~~~~~~~~~~~~~~~~~~ 1 warning generated when compiling for host. [ 91%] Building CXX object src/CMakeFiles/llama.dir/unicode-data.cpp.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/src && /usr/bin/hipcc -DGGML_BACKEND_SHARED -DGGML_SHARED -DGGML_USE_CPU -DGGML_USE_CUDA -DGGML_USE_HIP -DLLAMA_BUILD -DLLAMA_SHARED -Dllama_EXPORTS -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/../include -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -fPIC -MD -MT src/CMakeFiles/llama.dir/unicode-data.cpp.o -MF CMakeFiles/llama.dir/unicode-data.cpp.o.d -o CMakeFiles/llama.dir/unicode-data.cpp.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/unicode-data.cpp sh: line 1: /usr/bin/rocm_agent_enumerator: No such file or directory 1 warning generated when compiling for host. [ 92%] Linking CXX shared library libllama.so cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/src && /usr/bin/cmake -E cmake_link_script CMakeFiles/llama.dir/link.txt --verbose=1 sh: line 1: /usr/bin/rocm_agent_enumerator: No such file or directory clang++: warning: argument unused during compilation: '-Xarch_host -fstack-protector-strong' [-Wunused-command-line-argument] clang++: warning: argument unused during compilation: '-Xarch_host -fcf-protection' [-Wunused-command-line-argument] clang++: warning: argument unused during compilation: '-specs=/usr/lib/rpm/redhat/redhat-package-notes' [-Wunused-command-line-argument] /usr/bin/hipcc -fPIC -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -Xlinker --dependency-file=CMakeFiles/llama.dir/link.d -Wl,-z,relro -Wl,--as-needed -Wl,-z,pack-relative-relocs -Wl,-z,now -Wl,-z,now -Wl,--build-id=sha1 -specs=/usr/lib/rpm/redhat/redhat-package-notes -shared -Wl,-soname,libllama.so.b4094 -o libllama.so.b4094 CMakeFiles/llama.dir/llama.cpp.o "CMakeFiles/llama.dir/llama-vocab.cpp.o" "CMakeFiles/llama.dir/llama-grammar.cpp.o" "CMakeFiles/llama.dir/llama-sampling.cpp.o" CMakeFiles/llama.dir/unicode.cpp.o "CMakeFiles/llama.dir/unicode-data.cpp.o" ../ggml/src/libggml.so.b4094 ../ggml/src/ggml-cpu/libggml-cpu.so ../ggml/src/ggml-hip/libggml-hip.so ../ggml/src/libggml-base.so.b4094 cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/src && /usr/bin/cmake -E cmake_symlink_library libllama.so.b4094 libllama.so.b4094 libllama.so gmake[2]: Leaving directory '/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build' [ 92%] Built target llama /usr/bin/gmake -f common/CMakeFiles/common.dir/build.make common/CMakeFiles/common.dir/depend gmake[2]: Entering directory '/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build' cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build && /usr/bin/cmake -E cmake_depends "Unix Makefiles" /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094 /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/common /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/common /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/common/CMakeFiles/common.dir/DependInfo.cmake "--color=" gmake[2]: Leaving directory '/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build' /usr/bin/gmake -f common/CMakeFiles/common.dir/build.make common/CMakeFiles/common.dir/build gmake[2]: Entering directory '/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build' [ 93%] Building CXX object common/CMakeFiles/common.dir/arg.cpp.o [ 94%] Building CXX object common/CMakeFiles/common.dir/common.cpp.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/common && /usr/bin/hipcc -DGGML_BACKEND_SHARED -DGGML_SHARED -DGGML_USE_CPU -DGGML_USE_CUDA -DGGML_USE_HIP -DLLAMA_SHARED -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/common/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/../include -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -fPIC -MD -MT common/CMakeFiles/common.dir/arg.cpp.o -MF CMakeFiles/common.dir/arg.cpp.o.d -o CMakeFiles/common.dir/arg.cpp.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/common/arg.cpp cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/common && /usr/bin/hipcc -DGGML_BACKEND_SHARED -DGGML_SHARED -DGGML_USE_CPU -DGGML_USE_CUDA -DGGML_USE_HIP -DLLAMA_SHARED -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/common/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/../include -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -fPIC -MD -MT common/CMakeFiles/common.dir/common.cpp.o -MF CMakeFiles/common.dir/common.cpp.o.d -o CMakeFiles/common.dir/common.cpp.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/common/common.cpp sh: line 1: /usr/bin/rocm_agent_enumerator: No such file or directory sh: line 1: /usr/bin/rocm_agent_enumerator: No such file or directory In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/common/common.cpp:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/common/common.h:410:26: warning: unused function 'string_split>' [-Wunused-function] 410 | std::vector string_split(const std::string & input, char separator) | ^~~~~~~~~~~~~~~~~~~~~~~~~ 1 warning generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/common/common.cpp:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/common/common.h:410:26: warning: unused function 'string_split>' [-Wunused-function] 410 | std::vector string_split(const std::string & input, char separator) | ^~~~~~~~~~~~~~~~~~~~~~~~~ 1 warning generated when compiling for host. [ 95%] Building CXX object common/CMakeFiles/common.dir/console.cpp.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/common && /usr/bin/hipcc -DGGML_BACKEND_SHARED -DGGML_SHARED -DGGML_USE_CPU -DGGML_USE_CUDA -DGGML_USE_HIP -DLLAMA_SHARED -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/common/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/../include -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -fPIC -MD -MT common/CMakeFiles/common.dir/console.cpp.o -MF CMakeFiles/common.dir/console.cpp.o.d -o CMakeFiles/common.dir/console.cpp.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/common/console.cpp sh: line 1: /usr/bin/rocm_agent_enumerator: No such file or directory [ 96%] Building CXX object common/CMakeFiles/common.dir/json-schema-to-grammar.cpp.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/common && /usr/bin/hipcc -DGGML_BACKEND_SHARED -DGGML_SHARED -DGGML_USE_CPU -DGGML_USE_CUDA -DGGML_USE_HIP -DLLAMA_SHARED -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/common/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/../include -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -fPIC -MD -MT common/CMakeFiles/common.dir/json-schema-to-grammar.cpp.o -MF CMakeFiles/common.dir/json-schema-to-grammar.cpp.o.d -o CMakeFiles/common.dir/json-schema-to-grammar.cpp.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/common/json-schema-to-grammar.cpp sh: line 1: /usr/bin/rocm_agent_enumerator: No such file or directory [ 97%] Building CXX object common/CMakeFiles/common.dir/log.cpp.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/common && /usr/bin/hipcc -DGGML_BACKEND_SHARED -DGGML_SHARED -DGGML_USE_CPU -DGGML_USE_CUDA -DGGML_USE_HIP -DLLAMA_SHARED -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/common/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/../include -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -fPIC -MD -MT common/CMakeFiles/common.dir/log.cpp.o -MF CMakeFiles/common.dir/log.cpp.o.d -o CMakeFiles/common.dir/log.cpp.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/common/log.cpp sh: line 1: /usr/bin/rocm_agent_enumerator: No such file or directory [ 98%] Building CXX object common/CMakeFiles/common.dir/ngram-cache.cpp.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/common && /usr/bin/hipcc -DGGML_BACKEND_SHARED -DGGML_SHARED -DGGML_USE_CPU -DGGML_USE_CUDA -DGGML_USE_HIP -DLLAMA_SHARED -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/common/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/../include -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -fPIC -MD -MT common/CMakeFiles/common.dir/ngram-cache.cpp.o -MF CMakeFiles/common.dir/ngram-cache.cpp.o.d -o CMakeFiles/common.dir/ngram-cache.cpp.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/common/ngram-cache.cpp sh: line 1: /usr/bin/rocm_agent_enumerator: No such file or directory In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/common/ngram-cache.cpp:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/common/common.h:410:26: warning: unused function 'string_split>' [-Wunused-function] 410 | std::vector string_split(const std::string & input, char separator) | ^~~~~~~~~~~~~~~~~~~~~~~~~ 1 warning generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/common/ngram-cache.cpp:2: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/common/common.h:410:26: warning: unused function 'string_split>' [-Wunused-function] 410 | std::vector string_split(const std::string & input, char separator) | ^~~~~~~~~~~~~~~~~~~~~~~~~ 1 warning generated when compiling for host. [ 99%] Building CXX object common/CMakeFiles/common.dir/sampling.cpp.o cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/common && /usr/bin/hipcc -DGGML_BACKEND_SHARED -DGGML_SHARED -DGGML_USE_CPU -DGGML_USE_CUDA -DGGML_USE_HIP -DLLAMA_SHARED -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/common/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/. -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/src/../include -I/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/ggml/src/../include -O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -DNDEBUG -fPIC -MD -MT common/CMakeFiles/common.dir/sampling.cpp.o -MF CMakeFiles/common.dir/sampling.cpp.o.d -o CMakeFiles/common.dir/sampling.cpp.o -c /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/common/sampling.cpp sh: line 1: /usr/bin/rocm_agent_enumerator: No such file or directory In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/common/sampling.cpp:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/common/sampling.h:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/common/common.h:410:26: warning: unused function 'string_split>' [-Wunused-function] 410 | std::vector string_split(const std::string & input, char separator) | ^~~~~~~~~~~~~~~~~~~~~~~~~ 1 warning generated when compiling for gfx906. In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/common/sampling.cpp:1: In file included from /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/common/sampling.h:5: /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/common/common.h:410:26: warning: unused function 'string_split>' [-Wunused-function] 410 | std::vector string_split(const std::string & input, char separator) | ^~~~~~~~~~~~~~~~~~~~~~~~~ 1 warning generated when compiling for host. [100%] Linking CXX static library libcommon.a cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/common && /usr/bin/cmake -P CMakeFiles/common.dir/cmake_clean_target.cmake cd /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/common && /usr/bin/cmake -E cmake_link_script CMakeFiles/common.dir/link.txt --verbose=1 bfd plugin: LLVM gold plugin has failed to create LTO module: Invalid record bfd plugin: LLVM gold plugin has failed to create LTO module: Invalid record bfd plugin: LLVM gold plugin has failed to create LTO module: Invalid record bfd plugin: LLVM gold plugin has failed to create LTO module: Invalid record bfd plugin: LLVM gold plugin has failed to create LTO module: Invalid record bfd plugin: LLVM gold plugin has failed to create LTO module: Invalid record bfd plugin: LLVM gold plugin has failed to create LTO module: Invalid record bfd plugin: LLVM gold plugin has failed to create LTO module: Invalid record bfd plugin: LLVM gold plugin has failed to create LTO module: Invalid record bfd plugin: LLVM gold plugin has failed to create LTO module: Invalid record bfd plugin: LLVM gold plugin has failed to create LTO module: Invalid record bfd plugin: LLVM gold plugin has failed to create LTO module: Invalid record bfd plugin: LLVM gold plugin has failed to create LTO module: Invalid record bfd plugin: LLVM gold plugin has failed to create LTO module: Invalid record bfd plugin: LLVM gold plugin has failed to create LTO module: Invalid record bfd plugin: LLVM gold plugin has failed to create LTO module: Invalid record /usr/bin/ar qc libcommon.a CMakeFiles/common.dir/arg.cpp.o CMakeFiles/common.dir/common.cpp.o CMakeFiles/common.dir/console.cpp.o "CMakeFiles/common.dir/json-schema-to-grammar.cpp.o" CMakeFiles/common.dir/log.cpp.o "CMakeFiles/common.dir/ngram-cache.cpp.o" CMakeFiles/common.dir/sampling.cpp.o "CMakeFiles/build_info.dir/build-info.cpp.o" /usr/bin/ranlib libcommon.a gmake[2]: Leaving directory '/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build' [100%] Built target common gmake[1]: Leaving directory '/builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build' /usr/bin/cmake -E cmake_progress_start /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/redhat-linux-build/CMakeFiles 0 + module purge + local _mlredir=0 + '[' -n '' ']' + case " $@ " in + '[' 0 -eq 0 ']' + _module_raw purge ++ /usr/bin/tclsh /usr/share/Modules/libexec/modulecmd.tcl bash purge + eval 'unset __MODULES_LMCONFLICT; unset ROCM_BIN; unset _LMFILES_; unset LOADEDMODULES; unset ROCM_GPUS; unset ROCM_LIB; test 0;' ++ unset __MODULES_LMCONFLICT ++ unset ROCM_BIN ++ unset _LMFILES_ ++ unset LOADEDMODULES ++ unset ROCM_GPUS ++ unset ROCM_LIB ++ test 0 + _mlstatus=0 + return 0 + RPM_EC=0 ++ jobs -p + exit 0 Executing(%install): /bin/sh -e /var/tmp/rpm-tmp.kcq3oA + umask 022 + cd /builddir/build/BUILD/llama-cpp-b4094-build + '[' /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT '!=' / ']' + rm -rf /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT ++ dirname /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT + mkdir -p /builddir/build/BUILD/llama-cpp-b4094-build + mkdir /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT + CFLAGS='-O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer ' + export CFLAGS + CXXFLAGS='-O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -Xarch_host -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -Xarch_host -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer' + export CXXFLAGS + FFLAGS='-O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -I/usr/lib64/gfortran/modules ' + export FFLAGS + FCFLAGS='-O2 -flto=thin -fexceptions -g -grecord-gcc-switches -pipe -Wall -Wp,-U_FORTIFY_SOURCE,-D_FORTIFY_SOURCE=3 -Wp,-D_GLIBCXX_ASSERTIONS --config /usr/lib/rpm/redhat/redhat-hardened-clang.cfg -fstack-protector-strong -m64 -march=x86-64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -I/usr/lib64/gfortran/modules ' + export FCFLAGS + VALAFLAGS=-g + export VALAFLAGS + RUSTFLAGS='-Copt-level=3 -Cdebuginfo=2 -Ccodegen-units=1 -Cstrip=none -Cforce-frame-pointers=yes -Clink-arg=-specs=/usr/lib/rpm/redhat/redhat-package-notes --cap-lints=warn' + export RUSTFLAGS + LDFLAGS='-Wl,-z,relro -Wl,--as-needed -Wl,-z,pack-relative-relocs -Wl,-z,now -Wl,-z,now -Wl,--build-id=sha1 -specs=/usr/lib/rpm/redhat/redhat-package-notes ' + export LDFLAGS + LT_SYS_LIBRARY_PATH=/usr/lib64: + export LT_SYS_LIBRARY_PATH + CC=hipcc + export CC + CXX=hipcc + export CXX + cd llama.cpp-b4094 + DESTDIR=/builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT + /usr/bin/cmake --install redhat-linux-build -- Install configuration: "Release" -- Installing: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/lib64/libggml-cpu.so -- Installing: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/lib64/libggml-hip.so -- Installing: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/lib64/libggml.so.b4094 -- Installing: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/lib64/libggml.so -- Installing: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/include/ggml.h -- Installing: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/include/ggml-cpu.h -- Installing: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/include/ggml-alloc.h -- Installing: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/include/ggml-backend.h -- Installing: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/include/ggml-blas.h -- Installing: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/include/ggml-cann.h -- Installing: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/include/ggml-cuda.h -- Installing: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/include/ggml-kompute.h -- Installing: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/include/ggml-metal.h -- Installing: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/include/ggml-rpc.h -- Installing: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/include/ggml-sycl.h -- Installing: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/include/ggml-vulkan.h -- Up-to-date: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/lib64/libggml.so.b4094 -- Up-to-date: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/lib64/libggml.so -- Up-to-date: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/include/ggml.h -- Up-to-date: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/include/ggml-cpu.h -- Up-to-date: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/include/ggml-alloc.h -- Up-to-date: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/include/ggml-backend.h -- Up-to-date: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/include/ggml-blas.h -- Up-to-date: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/include/ggml-cann.h -- Up-to-date: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/include/ggml-cuda.h -- Up-to-date: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/include/ggml-kompute.h -- Up-to-date: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/include/ggml-metal.h -- Up-to-date: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/include/ggml-rpc.h -- Up-to-date: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/include/ggml-sycl.h -- Up-to-date: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/include/ggml-vulkan.h -- Installing: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/lib64/libggml-base.so.b4094 -- Installing: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/lib64/libggml-base.so -- Installing: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/lib64/libllama.so.b4094 -- Installing: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/lib64/libllama.so -- Installing: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/include/llama.h -- Installing: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/lib64/cmake/llama/llama-config.cmake -- Installing: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/lib64/cmake/llama/llama-version.cmake -- Installing: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/bin/convert_hf_to_gguf.py -- Installing: /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/lib/pkgconfig/llama.pc + rm -rf '/builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/lib64/libggml_shared.*' + rm /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/bin/convert_hf_to_gguf.py + /usr/bin/find-debuginfo -j2 --strict-build-id -m -i --build-id-seed b4094-8.fc42 --unique-debug-suffix -b4094-8.fc42.x86_64 --unique-debug-src-base llama-cpp-b4094-8.fc42.x86_64 --run-dwz --dwz-low-mem-die-limit 10000000 --dwz-max-die-limit 110000000 -S debugsourcefiles.list /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094 find-debuginfo: starting Extracting debug info from 5 files DWARF-compressing 5 files dwz: ./usr/lib64/libggml-base.so.b4094-b4094-8.fc42.x86_64.debug: Unknown debugging section .debug_str_offsets dwz: ./usr/lib64/libggml-cpu.so-b4094-8.fc42.x86_64.debug: Unknown debugging section .debug_str_offsets dwz: ./usr/lib64/libggml-hip.so-b4094-8.fc42.x86_64.debug: Unknown debugging section .debug_str_offsets dwz: ./usr/lib64/libggml.so.b4094-b4094-8.fc42.x86_64.debug: Unknown debugging section .debug_str_offsets dwz: ./usr/lib64/libllama.so.b4094-b4094-8.fc42.x86_64.debug: Unknown debugging section .debug_str_offsets dwz: Too few files for multifile optimization sepdebugcrcfix: Updated 0 CRC32s, 5 CRC32s did match. Creating .debug symlinks for symlinks to ELF files Copying sources found by 'debugedit -l' to /usr/src/debug/llama-cpp-b4094-8.fc42.x86_64 find-debuginfo: done + /usr/lib/rpm/check-buildroot + /usr/lib/rpm/redhat/brp-ldconfig + /usr/lib/rpm/brp-compress + /usr/lib/rpm/redhat/brp-strip-lto /usr/bin/strip + /usr/lib/rpm/brp-strip-static-archive /usr/bin/strip + /usr/lib/rpm/check-rpaths + /usr/lib/rpm/redhat/brp-mangle-shebangs + /usr/lib/rpm/brp-remove-la-files + env /usr/lib/rpm/redhat/brp-python-bytecompile '' 1 0 -j2 + /usr/lib/rpm/redhat/brp-python-hardlink + /usr/bin/add-determinism --brp -j2 /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT Scanned 30 directories and 168 files, processed 0 inodes, 0 modified (0 replaced + 0 rewritten), 0 unsupported format, 0 errors Reading /builddir/build/BUILD/llama-cpp-b4094-build/SPECPARTS/rpm-debuginfo.specpart Processing files: llama-cpp-b4094-8.fc42.x86_64 Executing(%license): /bin/sh -e /var/tmp/rpm-tmp.3JnJdW + umask 022 + cd /builddir/build/BUILD/llama-cpp-b4094-build + cd llama.cpp-b4094 + LICENSEDIR=/builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/share/licenses/llama-cpp + export LC_ALL=C.UTF-8 + LC_ALL=C.UTF-8 + export LICENSEDIR + /usr/bin/mkdir -p /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/share/licenses/llama-cpp + cp -pr /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/LICENSE /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/share/licenses/llama-cpp + RPM_EC=0 ++ jobs -p + exit 0 Provides: libggml-base.so.b4094()(64bit) libggml.so.b4094()(64bit) libllama.so.b4094()(64bit) llama-cpp = b4094-8.fc42 llama-cpp(x86-64) = b4094-8.fc42 Requires(rpmlib): rpmlib(CompressedFileNames) <= 3.0.4-1 rpmlib(FileDigests) <= 4.6.0-1 rpmlib(PayloadFilesHavePrefix) <= 4.0-1 Requires: ld-linux-x86-64.so.2()(64bit) ld-linux-x86-64.so.2(GLIBC_2.3)(64bit) libc.so.6()(64bit) libc.so.6(GLIBC_2.14)(64bit) libc.so.6(GLIBC_2.17)(64bit) libc.so.6(GLIBC_2.2.5)(64bit) libc.so.6(GLIBC_2.3.4)(64bit) libc.so.6(GLIBC_2.32)(64bit) libc.so.6(GLIBC_2.34)(64bit) libc.so.6(GLIBC_2.38)(64bit) libc.so.6(GLIBC_2.4)(64bit) libc.so.6(GLIBC_ABI_DT_RELR)(64bit) libgcc_s.so.1()(64bit) libgcc_s.so.1(GCC_3.0)(64bit) libggml-base.so.b4094()(64bit) libggml-cpu.so()(64bit) libggml-hip.so()(64bit) libggml.so.b4094()(64bit) libm.so.6()(64bit) libm.so.6(GLIBC_2.2.5)(64bit) libm.so.6(GLIBC_2.27)(64bit) libm.so.6(GLIBC_2.29)(64bit) libstdc++.so.6()(64bit) libstdc++.so.6(CXXABI_1.3)(64bit) libstdc++.so.6(CXXABI_1.3.11)(64bit) libstdc++.so.6(CXXABI_1.3.13)(64bit) libstdc++.so.6(CXXABI_1.3.2)(64bit) libstdc++.so.6(CXXABI_1.3.3)(64bit) libstdc++.so.6(CXXABI_1.3.5)(64bit) libstdc++.so.6(GLIBCXX_3.4)(64bit) libstdc++.so.6(GLIBCXX_3.4.11)(64bit) libstdc++.so.6(GLIBCXX_3.4.14)(64bit) libstdc++.so.6(GLIBCXX_3.4.15)(64bit) libstdc++.so.6(GLIBCXX_3.4.17)(64bit) libstdc++.so.6(GLIBCXX_3.4.18)(64bit) libstdc++.so.6(GLIBCXX_3.4.19)(64bit) libstdc++.so.6(GLIBCXX_3.4.20)(64bit) libstdc++.so.6(GLIBCXX_3.4.21)(64bit) libstdc++.so.6(GLIBCXX_3.4.22)(64bit) libstdc++.so.6(GLIBCXX_3.4.25)(64bit) libstdc++.so.6(GLIBCXX_3.4.26)(64bit) libstdc++.so.6(GLIBCXX_3.4.29)(64bit) libstdc++.so.6(GLIBCXX_3.4.30)(64bit) libstdc++.so.6(GLIBCXX_3.4.9)(64bit) Recommends: numactl Processing files: llama-cpp-devel-b4094-8.fc42.x86_64 Executing(%doc): /bin/sh -e /var/tmp/rpm-tmp.2Hnl0E + umask 022 + cd /builddir/build/BUILD/llama-cpp-b4094-build + cd llama.cpp-b4094 + DOCDIR=/builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/share/doc/llama-cpp-devel + export LC_ALL=C.UTF-8 + LC_ALL=C.UTF-8 + export DOCDIR + /usr/bin/mkdir -p /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/share/doc/llama-cpp-devel + cp -pr /builddir/build/BUILD/llama-cpp-b4094-build/llama.cpp-b4094/README.md /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT/usr/share/doc/llama-cpp-devel + RPM_EC=0 ++ jobs -p + exit 0 Provides: cmake(llama) libggml-cpu.so()(64bit) libggml-hip.so()(64bit) llama-cpp-devel = b4094-8.fc42 llama-cpp-devel(x86-64) = b4094-8.fc42 Requires(rpmlib): rpmlib(CompressedFileNames) <= 3.0.4-1 rpmlib(FileDigests) <= 4.6.0-1 rpmlib(PayloadFilesHavePrefix) <= 4.0-1 Requires: cmake-filesystem(x86-64) libamdhip64.so.6()(64bit) libamdhip64.so.6(hip_4.2)(64bit) libamdhip64.so.6(hip_6.0)(64bit) libc.so.6()(64bit) libc.so.6(GLIBC_2.14)(64bit) libc.so.6(GLIBC_2.2.5)(64bit) libc.so.6(GLIBC_2.29)(64bit) libc.so.6(GLIBC_2.3.2)(64bit) libc.so.6(GLIBC_2.3.4)(64bit) libc.so.6(GLIBC_2.32)(64bit) libc.so.6(GLIBC_2.33)(64bit) libc.so.6(GLIBC_2.34)(64bit) libc.so.6(GLIBC_2.4)(64bit) libc.so.6(GLIBC_2.7)(64bit) libc.so.6(GLIBC_ABI_DT_RELR)(64bit) libgcc_s.so.1()(64bit) libgcc_s.so.1(GCC_3.0)(64bit) libggml-base.so.b4094()(64bit) libggml.so.b4094()(64bit) libhipblas.so.2()(64bit) libllama.so.b4094()(64bit) libm.so.6()(64bit) libm.so.6(GLIBC_2.2.5)(64bit) libm.so.6(GLIBC_2.27)(64bit) libm.so.6(GLIBC_2.29)(64bit) librocblas.so.4()(64bit) libstdc++.so.6()(64bit) libstdc++.so.6(CXXABI_1.3)(64bit) libstdc++.so.6(GLIBCXX_3.4)(64bit) libstdc++.so.6(GLIBCXX_3.4.11)(64bit) libstdc++.so.6(GLIBCXX_3.4.30)(64bit) Processing files: llama-cpp-debugsource-b4094-8.fc42.x86_64 Provides: llama-cpp-debugsource = b4094-8.fc42 llama-cpp-debugsource(x86-64) = b4094-8.fc42 Requires(rpmlib): rpmlib(CompressedFileNames) <= 3.0.4-1 rpmlib(FileDigests) <= 4.6.0-1 rpmlib(PayloadFilesHavePrefix) <= 4.0-1 Processing files: llama-cpp-debuginfo-b4094-8.fc42.x86_64 Provides: debuginfo(build-id) = 51fbd24a455569cc482df3e73077bb74cb045489 debuginfo(build-id) = 66698019d8badba76ee7867a6b189dce5acff2c1 debuginfo(build-id) = 924fd4675f3b9adcba0ae82cbeb95fca92ef108f libggml-base.so.b4094-b4094-8.fc42.x86_64.debug()(64bit) libggml.so.b4094-b4094-8.fc42.x86_64.debug()(64bit) libllama.so.b4094-b4094-8.fc42.x86_64.debug()(64bit) llama-cpp-debuginfo = b4094-8.fc42 llama-cpp-debuginfo(x86-64) = b4094-8.fc42 Requires(rpmlib): rpmlib(CompressedFileNames) <= 3.0.4-1 rpmlib(FileDigests) <= 4.6.0-1 rpmlib(PayloadFilesHavePrefix) <= 4.0-1 Recommends: llama-cpp-debugsource(x86-64) = b4094-8.fc42 Processing files: llama-cpp-devel-debuginfo-b4094-8.fc42.x86_64 Provides: debuginfo(build-id) = 39218d62f5eac696ee4e97235d499baf110d873e debuginfo(build-id) = 6aac7771b17b855d3649070158d5af598901ea8f libggml-cpu.so-b4094-8.fc42.x86_64.debug()(64bit) libggml-hip.so-b4094-8.fc42.x86_64.debug()(64bit) llama-cpp-devel-debuginfo = b4094-8.fc42 llama-cpp-devel-debuginfo(x86-64) = b4094-8.fc42 Requires(rpmlib): rpmlib(CompressedFileNames) <= 3.0.4-1 rpmlib(FileDigests) <= 4.6.0-1 rpmlib(PayloadFilesHavePrefix) <= 4.0-1 Recommends: llama-cpp-debugsource(x86-64) = b4094-8.fc42 Checking for unpackaged file(s): /usr/lib/rpm/check-files /builddir/build/BUILD/llama-cpp-b4094-build/BUILDROOT Wrote: /builddir/build/RPMS/llama-cpp-debuginfo-b4094-8.fc42.x86_64.rpm Wrote: /builddir/build/RPMS/llama-cpp-devel-debuginfo-b4094-8.fc42.x86_64.rpm Wrote: /builddir/build/RPMS/llama-cpp-debugsource-b4094-8.fc42.x86_64.rpm Wrote: /builddir/build/RPMS/llama-cpp-b4094-8.fc42.x86_64.rpm Wrote: /builddir/build/RPMS/llama-cpp-devel-b4094-8.fc42.x86_64.rpm Executing(rmbuild): /bin/sh -e /var/tmp/rpm-tmp.P5K2fk + umask 022 + cd /builddir/build/BUILD/llama-cpp-b4094-build + test -d /builddir/build/BUILD/llama-cpp-b4094-build + /usr/bin/chmod -Rf a+rX,u+w,g-w,o-w /builddir/build/BUILD/llama-cpp-b4094-build + rm -rf /builddir/build/BUILD/llama-cpp-b4094-build + RPM_EC=0 ++ jobs -p + exit 0 Finish: rpmbuild llama-cpp-b4094-8.fc42.src.rpm Finish: build phase for llama-cpp-b4094-8.fc42.src.rpm INFO: chroot_scan: 1 files copied to /var/lib/copr-rpmbuild/results/chroot_scan INFO: /var/lib/mock/fedora-rawhide-x86_64-1736126189.030111/root/var/log/dnf5.log INFO: chroot_scan: creating tarball /var/lib/copr-rpmbuild/results/chroot_scan.tar.gz /bin/tar: Removing leading `/' from member names INFO: Done(/var/lib/copr-rpmbuild/results/llama-cpp-b4094-8.fc42.src.rpm) Config(child) 112 minutes 32 seconds INFO: Results and/or logs in: /var/lib/copr-rpmbuild/results INFO: Cleaning up build root ('cleanup_on_success=True') Start: clean chroot INFO: unmounting tmpfs. Finish: clean chroot Finish: run Running RPMResults tool Package info: { "packages": [ { "name": "llama-cpp-devel", "epoch": null, "version": "b4094", "release": "8.fc42", "arch": "x86_64" }, { "name": "llama-cpp-debugsource", "epoch": null, "version": "b4094", "release": "8.fc42", "arch": "x86_64" }, { "name": "llama-cpp", "epoch": null, "version": "b4094", "release": "8.fc42", "arch": "x86_64" }, { "name": "llama-cpp-debuginfo", "epoch": null, "version": "b4094", "release": "8.fc42", "arch": "x86_64" }, { "name": "llama-cpp-devel-debuginfo", "epoch": null, "version": "b4094", "release": "8.fc42", "arch": "x86_64" }, { "name": "llama-cpp", "epoch": null, "version": "b4094", "release": "8.fc42", "arch": "src" } ] } RPMResults finished